Busin3ss

THE RELEVANCE OF MISSION STATEMENTS Analysing the antecedents of perceived message quality and its relationship to employee mission engagement

Sebastian Desmidt

Sebastian Desmidt Faculty of Economics and Business Administration Ghent University Ghent Belgium E-mail: [email protected]

Abstract

Although mission statements are deemed effective tools for communicating the goals of public organizations, there is a dearth of research examining how employees perceive mission statements and their effect. Data from 1,418 employees of a Belgian public organiza- tion indicate that although perceived mission statement quality and employee mission engagement are positively related, individual acceptance of the mission statement varies within the analysed organization and can be, partially, explained by cognitions and attributes of the message receiver (hierarchical position, perceived self-efficacy and person–organiza- tion fit), and employee cognitions regarding the message sender (behavioural integrity) and the message (mission ambiguity).

Key words mission statement, employee perceptions, employee mission engagement

Public Management Review, 2016

Vol. 18, No. 6, 894–917, http://dx.doi.org/10.1080/14719037.2015.1051573

© 2015 Taylor & Francis

INTRODUCTION

Public management scholars have long noticed that public organizations are often characterized by goal ambiguity, which can have dysfunctional effects on employees’ affective responses towards the organization. More specifically, research indicated that goal ambiguity is negatively related to managerial effectiveness (Chun and Rainey 2005b) and role ambiguity (Pandey and Garnett 2006), whereas its counterpart, goal clarity, has been shown to have a positive relationship with employee work motivation (Jung and Rainey 2011), goal commitment (Jung 2013), communication performance (Pandey and Garnett 2006) and mission valence (Wright and Pandey 2011). To counterbalance the detrimental effects of such goal ambiguity, public organizations

have been searching for tools that would help them to ‘relentlessly communicate the organization’s mission in ways that not only clearly articulate what the organization hopes to accomplish, but also how the organization hopes to accomplish it and why such accomplishments benefit the community it services’ (Wright, Moynihan, and Pandey 2012, 212). One of these tools, which has been widely adopted by the public sector, is the process of formulating a mission statement (Weiss and Piderit 1999). In most cases, these mission statements are crafted with the intention that they will both serve as blueprints for what the organization stands for and create a shared interpretation of the organiza- tion’s essence by raising employee awareness of organizational values and goals (Paarlberg and Lavigna 2010). Moreover, the extant literature on the topic (Desmidt, Prinzie, and Decramer 2011) suggests that a specific articulation of what the organization represents will act as an ‘ideological and cognitive framework that connects followers identities to the collective identity of their organization, thereby increasing followers’ intrinsic motivation (rather than just providing extrinsic motivation) to perform their job’ (Park and Rainey 2008, 111). Accordingly, the act of communicating a mission statement can be interpreted as a managerial attempt to influence employee cognitions, attitudes and behaviour by disseminating specific labels to describe the organization, with the intention of stimulating a collective commitment to action (Maitlis and Lawrence 2007). However, although the popularity of mission statements is largely based on the

postulate that they are an effective internal communication tool (Paarlberg and Lavigna 2010), this assumption has only sporadically been the subject of empirical analysis (Bartkus and Glassman 2008). In general, previous studies on the communication effectiveness of mission statements have addressed the topic from an actor’s perspective (Ravasi and Schultz 2006), focusing on exploring the who and the what: “Who are the actors engaged in mission statement formulation processes?”; “What is the content of mission statements?”; and “What are the linguistic characteristics of mission state- ments?” (Desmidt, Prinzie, and Decramer 2011). One could argue, from a traditional communication perspective, that the literature on the subject has primarily focused on analysing the source of the message and its content but has neglected to address the role of the message recipient and the effect of the message (e.g. belief, attitude, and/or

Desmidt: Employee mission statement perceptions 895

behaviour change) (Sonenshein and Dholakia 2012). This observation leads to the contradiction that although the effectiveness of mission statements is deemed contingent upon the extent to which they are supported by employees, there is a dearth of research examining how employees interpret such management-initiated persuasive messages (Shapiro 2013). A research gap which could be troublesome because employee under- standing and action are based upon individual interpretations of organizational informa- tion (Pandey and Garnett 2006). Consequently, the meaning of organizational messages, such as mission statements, is not only subjective but also constrained by the cognitions and attitudes of the employees experiencing them (Gioia and Chittipeddi 1991). This implies that

even employees in the same organization are not expected to interpret their [organization’s mission

statement] in identical ways as they will have different views on the clarity of the organization’s goals

[. . .] and the value of their organization’s mission or purpose. (Wright and Pandey 2011, 29)

This paper addresses the cited research gap by analysing how employee perceptions of a mission statement differ within an organization. More specifically, concepts and research methodologies from formative and communication research are used to assess, in the context of a Belgian public welfare organization (n = 1,418) (a) how the employees perceive the quality of their organization’s mission statement; (b) the relationship between perceived message quality and employee motivation to advance the organization’s mission and (c) how perceived source credibility (i.e. the behavioural integrity of organization leaders), perceived content clarity (i.e. mission comprehension ambiguity) and receiver characteristics (i.e. self-efficacy and person–organization fit (POF)) are able to explain intra-organizational variances in perceived message quality. The collected data were analysed using structural equation modelling.

HYPOTHESES

Arguing that mission statement communication can be viewed as a management- initiated communication activity, whereby mission statements are used as explicit narratives intended to influence employees’ perceptions of an organization’s central, enduring and distinctive features (Ravasi and Schultz 2006), highlights the assumption that mission statements are not only written to communicate an organization’s funda- mentals but also that ‘the act of communicating this mission message to certain parties is assumed to further the ends expressed in it’ (Hackley 1998, 94). Consequently, the communication effectiveness of mission statements can be interpreted as a function of their ability to influence the beliefs, assumptions and ultimately, behaviour of an organization’s employees. However, despite the widespread use of mission statements in the public sector (Weiss and Piderit 1999), the effectiveness of mission statements as

896 Public Management Review

a vehicle for organizational communication is underexploited (Smythe et al. 2006). As a result, mission statement research should focus more on developing methodologies to analyse not only how such management-initiated communication activities are perceived by employees but also their effect (Sonenshein and Dholakia 2012). One source of inspiration could be formative research. More specifically, formative research uses the concept of ‘perceived message quality’ to estimate ‘the degree to which a persuasive message will be favourably evaluated – in terms of its persuasive potential – by recipients of that message’ (Dillard, Weber, and Vail 2007, 614). In most cases, respondents are asked to read a message and then assess its persuasive quality by scoring it on specific features (e.g. plausible, compelling and sound). In addition, respondents are asked to provide information about their cognitive and affective responses to the message’s advocacy (Updegraff et al. 2007). The strength of the relationship between perceived message quality and an employee’s affective responses to the message’s advocacy is viewed as an indication of the message’s persuasive power. In addition, based on the theory of reasoned action, Fishbein et al. (2002) argue that an employee’s affective responses to a message’s advocacy are strong predictors of actual behaviour, because attitudes towards a given object will generally result in behaviours that are consistent with those attitudes.

Perceived message quality and employee mission engagement

Although the content, context and targeted audiences of mission statements and the subjects of traditional formative research (e.g. public service announcements and health messages) differ significantly, mission statement research could benefit from these methodologies, because they build on persuasive and motivational theories that are applicable at many levels, including intrapersonal, interpersonal and organizational (Cameron 2009). Accordingly, we argue that higher levels of perceived mission statement quality are positively related with employee’s intentions to help realize that mission statement. An employee’s affective attitude towards a mission statement’s advocacy is operationalized as ‘employee mission engagement’, which refers to ‘a psychological state in which an employee desires to exert effort and devote careful attention to ensure the fulfillment of a mission that he or she perceives as significant or meaningful’ (Suh et al. 2011, 78-79). Consequently, we hypothesize that:

Hypothesis 1: Perceived message quality is positively related to employee levels of mission engagement.

The selection of employee mission engagement as the dependent variable is in line with the requirements of formative research, which indicate that a respondent’s affective responses to a message’s advocacy should be measured using an operationaliza- tion that is closely linked to the content of the persuasive message under scrutiny (Updegraff et al. 2007). Because employee mission engagement specifically focuses on

Desmidt: Employee mission statement perceptions 897

individual intentions to advance the goals and raison d’être expressed in the mission statement, it seemed a more suitable operationalization than general measures of employee work attitudes (e.g. job satisfaction) or related constructs such as mission valence (which is developed to measure an ‘employee’s perceptions of the attractiveness or salience of an organization’s purpose or social contribution’ (Wright and Pandey 2011, 24)) and consequently does not ask employees to evaluate an organization’s mission statement (Wright, Moynihan, and Pandey 2012).

Antecedents of perceived message quality

In accordance with Wright, Moynihan, and Pandey’s (2012, 206) claim that ‘the long- standing recognition of the power of worthy and attractive goals [makes] it important to gain a better understanding of what makes organizational goals attractive and compel- ling to individual employees’, this paper attempts to uncover the mechanisms that make mission statements attractive to employees. Because multiple communication theories have indicated that the effectiveness of persuasive communication is predominantly influenced by characteristics of message content, message source and message receiver (Cameron 2009), we analyse how employee perceptions of content clarity (i.e. mission comprehension ambiguity), source credibility (i.e. the behavioural integrity of the organization’s leaders) and individual attributes of the message recipient (i.e. self- efficacy and POF) are related to perceived mission statement attractiveness. Accordingly, this study not only builds on the specifications of formative research but also on the underlying theories of behavioural prediction and motivation (e.g. Theory of Planned Behavior, Theory of Reasoned Action, and Social Cognitive Theory) to identify critical beliefs influencing a person’s intention to perform a specific behaviour or attitude change. More specifically, Fishbein and Cappella (2006, S2) argue that “these theories suggest that there are only a limited number of variables that must be considered in predicting and understanding any given behavior” or cognition, namely attitude towards the focal behaviour/cognition (conceptualized as the perceived degree of congruence between a recipient’s personal values and message advocacy or POF), perceived norms concerning the focal behaviour/cognition (conceptualized as perceived adherence to message advocacy by important referents or behavioural integrity), and self-efficacy with respect to the focal behaviour/cognition (conceptualized as one’s perceived ability to contribute to the realization of the message advocacy). First, we examine the relationship between mission comprehension ambiguity and

perceived message quality. Although public managers have the responsibility to articu- late a clear and compelling mission for the organization (Moynihan, Pandey, and Wright 2012), mission statements are often characterized by a high level of interpretive leeway. For example, a mission statement can focus on ‘delivering the highest possible quality’, but the labels ‘quality’ and ‘highest possible’ can have different meanings at different times or different places in the organization. High levels of such ‘mission

898 Public Management Review

comprehension ambiguity’, which refers ‘to the level of interpretive leeway that an organizational mission allows in comprehending, explaining, and communicating the organizational mission’ (Chun and Rainey 2005a, 531), have a detrimental impact on the motivational power and perceived attractiveness of organizational goals. Accordingly, we hypothesize as follows:

Hypothesis 2: Mission comprehension ambiguity is negatively related to perceived message quality.

Second, mission statements need to be more than a public relations exercise. Leaders wanting to encourage followers to transcend their own self-interests for the sake of larger organization goals, should both act as prosocial role models and exhibit sincere and sustained commitment to the projected organizational goals (Paarlberg and Lavigna 2010). If message recipients perceive inconsistencies between the content of persuasive messages and the actions of the message senders, the projected message will evoke feelings of scepticism (Narayanan, Zane, and Kemmerer 2011). Moynihan, Pandey, and Wright (2012) stress that successful leaders must use both words and deeds to inspire their employees, because employ- ees will do what they see their managers do rather than what policy manuals dictate (Paarlberg and Lavigna 2010). Simons et al. (2007) stress the importance of this managerial word–deed alignment and label this phenomenon as ‘behavioural integ- rity’, which entails the perceived fit between espoused and enacted values and includes employees’ perceptions of managers’ behavioural adherence to communica- tion instruments, such as mission statements and corporate value statements. A perceived lack of behavioural integrity can result in managers being accused of hypocrisy and can foster both cynicism and a loss of credibility, which will in turn decrease both the attractiveness and the validity of management communication (Moynihan, Pandey, and Wright 2012; Narayanan, Zane, and Kemmerer 2011; Urbany 2005). Subsequently, we argue the following:

Hypothesis 3: Behavioural integrity is positively related to perceived message quality.

A third factor that influences the level of perceived message quality is self-efficacy. Self-efficacy, which refers to one’s beliefs regarding one’s ability to perform a particular task or endeavour, is deemed to have a positive effect on the attitudinal and behavioural outcomes associated with goal accomplishment (Bandura 2012). More specifically, the concept of self-efficacy has been used to argue that employees with a clear under- standing of how their tasks fit the overall objectives and mission of an organization are more likely both to see the mission as meaningful and to display higher levels of dedication (Wright and Pandey 2011). Accordingly, we hypothesize as follows:

Hypothesis 4: Self-efficacy is positively related to perceived message quality.

Desmidt: Employee mission statement perceptions 899

A fourth factor that influences the effectiveness of a persuasive message is the existence of prior attitudes towards the message’s advocacy. More specifically, in line with the theory of reasoned action, the literature on persuasive messages demonstrates that message recipients will assess the value of a message by comparing it with their current attitudes (Cameron 2009). In essence,

messages that advocate positions within the receiver’s latitude of acceptance will be viewed as more

subjectively similar to the receiver’s position than is really the case. This assimilation effect is predicted

to produce more persuasion than would be expected from a realistic assessment of the message’s

position. The reverse or contrast effect holds when the message position falls into the receivers’ latitude

of rejection. (Dillard, Shen, and Vail 2007, 470)

As a result, prior attitudes can produce biased perceptions of persuasive messages and thus influence the persuasive power of those messages. With respect to the case of organizational missions, Wright, Moynihan, and Pandey (2012) use the theory of POF to claim that employee attitudes towards an organization’s mission are influenced by the level of perceived alignment between the employees’ personal values and those of the collective. Elaborating on this statement, we argue that the persuasive power of a value-laden instrument, such as a mission statement, is influenced by the perceived degree of congruency between the recipient’s personal values and the message advo- cated. Accordingly, we hypothesize as follows:

Hypothesis 5: Person–organization fit is positively related to perceived message quality.

Figure 1 provides an overview of the formulated hypotheses.

RESEARCH METHODS AND ANALYSIS

Research setting

The data for this study were collected in a Belgian public social service organization that had recently reformulated its mission statement. The reformulated statement was developed by a project team that comprised the organization’s top management team and members of the board of directors. The main goal of the mission reformulation process was to devise a more appealing mission statement using an active tone of voice. The reformulated statement thus differed from the previous statement in formulation and structure, whereas content and scope remained largely the same. Reformulating the mission statement did not imply a change in the strategic course of the organization, nor did it have immediate consequences for employees or their work.

900 Public Management Review

The reformulated mission statement is 243 words in length and specifies the purpose of the organization, the range of products and services offered, the primary service recipients and the created value for this stakeholder group, the geographic scope of the organization and the organization’s central values. The mission statement is character- ized by a Gunning-Fog Index score of 12.4, which indicates that it requires the reading level of a US high school senior. The reformulated mission statement was introduced to all employees via email and the organization’s newsletter, posted on the organization’s website, was visually displayed throughout the organization via posters, and included in an introduction brochure about the organization.

Data collection

The data were collected over a 2-week period 4 months after the initial commu- nication of the reformulated statement. More specifically, each employee (with the exception of the members of the top management team, because they were involved in the mission statement formulation process) was handed an envelope containing a copy of the organization’s mission statement, an eight-page questionnaire and a

Person- organization

fit

Behavioural integrity

Self-efficacy

Perceived message quality

Employee mission

engagement

Mission comprehension

ambiguity

H1

H2

H3

H4

H5

Figure 1: Hypothesized model

Desmidt: Employee mission statement perceptions 901

return envelope. Subjects had 2 weeks to return the completed questionnaires to a closed, on-site drop-off box. The survey had an N of 3,316, and 1,418 completed questionnaires were returned,

which resulted in a usable response rate of 46%. The respondents were primarily female (69.7%), with a mean age of 42 years (SD = 11), and they had been with the organization for 15 years on average (SD = 11.5). Fifty-six per cent of the respondents were employed full-time. Non-response bias was assessed by comparing the character- istics of the population (data available from company records whereby data were matched via a confidential code number) with the characteristics of the respondents. The comparison revealed no significant discrepancies between the samples and their respective populations with regards to gender, age, position, role or functional or organization tenure. A demographic overview of the entire organization’s workforce and the survey respondents is provided in Table 1.

Measures

The questionnaire contained forty-four closed-ended questions. All constructs except for message quality, which used a different semantic format, were measured using a seven-point Likert scale (ranging from strongly disagree (1) to strongly agree (7)). All items are validated measures from previous studies, although they were occasionally

Table 1: Characteristics of the survey respondents and the organization’s workforce

Respondents Organization (%)

Organization role Administration 15.5% 16.7% Technician/maintenance 3.5% 11.0% Professional 61.2% 56.1% Paraprofessional 19.8% 16.2%

Organization position Non-supervisor 86.5% 88.6% First-level manager 1.9% 8.9% Lower middle manager 1.5% 1.3% Higher middle manager 1.1% 1.2%

Gender Female 69.7% 72.2%

Mode of employment Full-time 55.7% 59.1%

Age (years) Avg. age 42 (SD = 11) 40 (SD = 12)

902 Public Management Review

slightly modified to enhance the focus on the organization’s mission statement (see Appendix 1 for an overview). Perceived message quality was measured by asking respondents to score the mission statement on specific quality features. The eight features were derived from Rainey and Steinbauer’s (1999) characteristics of mission valence (e.g. ‘Boring – Interesting’). Employee mission engagement was measured using a three-item scale developed by Suh et al. (2011) (e.g. ‘I am motivated by the mission statement to do my work’). Mission comprehension ambiguity was measured using a five-item scale. Three items were devised by Rainey (1983) and have been validated in various studies (Moynihan, Pandey, and Wright 2012; Pandey and Garnett 2006; Wright and Pandey 2011), and two items were developed by Fairhurst, Jordan, and Neuwirth (1997) (e.g. ‘The mission statement contains clearly defined goals’). Perceived behavioural integrity was measured with a three-item scale developed by Simons et al. (2007) (e.g. ‘The top management team of [organization’s name] practices what the mission statement preaches.’). Perceived self-efficacy was measured using a three-item scale developed by Wright and Pandey (2011) (e.g. ‘I have a clear understanding of how my work contributes to [organization’s name] mission state- ment.’), whereas perceived POF was measured using Cable, DeRue, and Scott’s (2002) three-item scale (e.g. ‘My personal values match the description of the organization’s values and culture in the mission statement.’). To isolate the influence of the hypothesized structural relationships, the variables of

age, gender, organization tenure, functional tenure, employment level, organization role and organization position were included as control variables.

Characteristics of the selected research design

This study relies on perceptual data collected through self-report questionnaires distributed in a single organization. Despite its advantages and being one of the most commonly used methods of data collection in public management research (Jakobsen and Jensen 2015; Favero and Bullock 2015), the selected research method has also inherent drawbacks that must be considered when interpreting the results. First, using a cross-sectional correlation design prevents the establishment of causal direction (Antonakis et al. 2014). Consequently, we avoid discussing causal inferences, because the suggested direction of the analysed relationships is theory-driven rather than the result of an observed temporal sequence. This also implies that reverse causality cannot be completely ruled out. Second, collecting data in only one organization creates an opportunity to analyse

intra-organizational variance in employee perceptions of organizational attributes but does not allow us to control for the impact of either organizational characteristics (e.g. communication climate, degree of formalization or organization size) or mission

Desmidt: Employee mission statement perceptions 903

statement properties (e.g. specificity, proximity, difficulty, complexity and/or conflict (Jung 2013)), which limits the opportunities to generalize the study findings. Third, ‘both the dependent and focal explanatory variables are perceptual measures

derived from the same respondent’ and therefore, common method variance (CMV) may be a concern (Chang, van Witteloostuijn, and Eden 2010, 178). More specifically,

in such cases, the estimated effect of one variable on another is at risk of being biased [by] systematic

variance shared among the variables, which is introduced to the measures by the measurement method

rather than the theoretical constructs the measures represent. (Jakobsen and Jensen 2015, 2)

Although the easiest way to prevent the occurrence of CMV is to measure both the predictor and criterion variables using different measurement methods (Podsakoff et al. 2003), this approach may not be the most appropriate to use when ‘both the predictor and criterion variables are capturing an individual’s perceptions, beliefs, judgments, or feelings’ (Podsakoff, MacKenzie, and Podsakoff 2012, 549). Moreover, when studying individual-level behaviours, attitudes, interpretations of events or behavioural intentions, self-reports are often the theoretically most relevant measurement method (Conway and Lance 2010; Meier and O’Toole 2013). Subsequently, because ‘even employees in the same organization [. . .] will have different views on the clarity of the organization’s goals [. . .] and the value of their organization’s mission or purpose’ (Wright and Pandey 2011, 29), using perceptual measures seems in this case to be justified. However, precautions have been taken to mitigate the potential impact of CMV. First, because common method bias is more likely to be a problem ‘when the

respondents can’t provide accurate responses (a function of their ability and the difficulty of the task) or when they are unwilling to try to provide accurate responses (which is a function of motivation)’ (Podsakoff, MacKenzie, and Podsakoff 2012, 560), as an ex ante measure, procedural remedies related to designing and administering the questionnaire were adopted (Chang, van Witteloostuijn, and Eden 2010). The issues of ability and difficulty were addressed using language, vocabulary and syntax that match the respondents’ reading capabilities (subject of a pretest), avoiding complex and abstract questions (subject of a pretest), labelling all response options rather than just the end points, varying the scale format, creating a methodological distance between specific criterion and predictor items using different formats (Likert scale versus semantic differential) and buffer items (MacKenzie and Podsakoff 2012; Podsakoff, MacKenzie, and Podsakoff 2012). A such, we tried to minimize the impact of two important potential sources of common method bias, i.e. common source bias (e.g. social desirability, consistency motif, positive and negative affectivity, respondents’ tendency to use (or avoid using) the extreme choices on a response scale) and bias produced by item characteristics (e.g. item complexity, ambiguity, or abstractness) (Jakobsen and Jensen 2015; Favero and Bullock 2015). The issue of the motivation to respond accurately was addressed by explaining to

respondents why the topic is important and how research can benefit the organization

904 Public Management Review

(cover letter), emphasizing that personal opinions are important and that there is no right or wrong answer (cover letter), avoiding lengthy scales, stressing the voluntary nature of participation and guaranteeing anonymity (MacKenzie and Podsakoff 2012; Podsakoff, MacKenzie, and Podsakoff 2012). Second, as an ex post measure, the single- common-method-factor approach was used to control for common method bias (Podsakoff et al. 2003). The addition of a method factor to a confirmatory factor model that includes the study variables as latent factors does not significantly improve the fit over the confirmatory factor model with just the hypothesized model’s latent factors (TLI = .045), although the variables factors’ loadings continued to be signifi- cant. Subsequently, the test results suggest that substantial common method bias is absent.

DATA ANALYSIS AND RESULTS

Prior to data analysis, missing values were replaced because testing the hypotheses using structural equation modelling assumes that the covariance matrix follows a Wishart distribution (Brown 1994). The missing values were imputed into IBM SPSS Statistics 21 (IBM Corp. Released 2013a) using EM missing value imputation given that they were neither missing completely at random (Little’s MCAR test χ2 = 2571.33, df = 2091, p = .000) nor missing at random.

Univariate and bivariate analysis

Table 2 provides the univariate and bivariate statistics for the study’s measures. The measures were reliable (Cronbach’s coefficient alphas ranging from .74 to .94 and construct reliabilities ranging from .75 to .96). In addition, the low-average bivariate correlation (.20) and the fact that no measure shared more than 42% of its variance with any other measure provide evidence of discriminant validity, whereas variance inflation factor tests suggested that multi-collinearity was not an issue (no score exceeded 1.7). With respect to the control variables, the statistics indicated that organization

tenure, functional tenure and organization role were significantly related to mission comprehension ambiguity. Organization position was significantly associated with all latent variables, indicating that the higher the employee’s position, the higher was the employee’s value for each of the latent variables except for mission comprehension ambiguity. Table 3 provides more details on this relationship. Because the means of all of the latent variables differed significantly depending on the

employee’s position, organization position was introduced into the structural model as an antecedent of the latent variables under investigation. Organization position was

Desmidt: Employee mission statement perceptions 905

Ta bl e 2:

D es cr ip tiv e st at is tic s fo r th e re se ar ch

co ns tr uc ts

M ea n

SD

Co rr el at io ns

an d co ns tr uc t re lia bi lit ie s in

pa re nt he se sa

1 2

3 4

5 6

7 8

9 10

11 12

13

St ud y Va ria bl es

1. Em

pl . m is si on

en ga ge m en t

4. 77

1. 19

(. 75 )

2. Pe rc ei ve d m es sa ge

qu al ity

4. 68

1. 15

.4 6* *

(. 94

)

3. M is si on

co m pr .

Am bi gu

ity 4. 75

1. 46

− .2 5* *

− .3 6* *

(. 92

)

4. Be ha vi ou

ra li nt eg rit y

4. 38

1. 20

.3 0* *

.4 8* *

− .4 1*

* (. 88 )

5. Se lf- ef fic ac y

4. 17

1. 31

.3 7* *

.5 **

− .2 4*

* .4 7* *

(. 76

) 6.

PO F

4. 88

1. 14

.5 8* *

.5 2* *

− .2 5*

* .4 3* *

.4 4* *

(. 85 )

Co nt ro lv ar ia bl es

7. G en de rb

.7 9

.4 1

.0 0

− .0 2

.0 2

− .0 1

− .0 3

− .0 1

N A

8. Ag

e 4. 0

11 .0

.0 1

.0 2

.0 1

.0 3

.0 0

.0 2

− .0 4

N A

9. O rg an iz at io n te nu re

14 .7

11 .0

.0 3

.0 2

− .1 3*

* .0 0

.0 3

.0 0

.0 4

.8 1*

** N A

10 . Fu nc tio na lt en ur e

12 .2

1. 1

.0 3

.0 2

− .0 9*

* .0 01

.0 3

.0 2

.0 8*

* .6 9*

** .8 3* **

N A

11 .M

od e of

em pl oy m en tc

.5 6

.5 0

.0 1

− .0 1

− .0 2

− .0 1

− .0 3

− .0 4

− .3 4*

** − .3 0*

** − .2 7* **

− .2 6*

** N A

12 . O rg an iz at io na l

po si tio nd

1. 00

.0 0

.1 4* **

.1 5* **

− .2 0*

** .1 6* **

.2 6* **

.1 4* **

.0 5

− .0 8*

* − .0 4

− .0 4

.0 1

N A

13 . O rg an iz at io na lr ol ee

1. 61

.8 7

.2 5

2. 33

12 .9 8*

** .1 6

1. 09

.2 3

13 .3 0*

** 7. 14 **

* 1. 12

1. 85

1. 51 **

* 8. 95 **

* N A

N ot es : a A ll ca lc ul at io ns

ar e Pe ar so n co rr el at io ns

ex ce pt

fo r O rg an iz at io n ro le fo r w hi ch

on e- w ay

AN O VA

F- st at is tic s (t es t of

di ffe re nc e)

w er e us ed .

b 0 = m al e;

1 = fe m al e.

c 0 = pa rt -t im e;

1 = fu ll- tim

e. d 1 = N on -s up er vi so r; 2 = fir st -le ve l m an ag er ; 3 = lo w er

m id dl e m an ag er ; 4 = hi gh er

m id dl e m an ag er .

e 1 = Pr of es si on al ; 2 = pa ra pr of es si on al ; 3 = ad m in is tr at io n;

4 = te ch ni ci an /m

ai nt en an ce .

** Co

rr el at io n is si gn ifi ca nt

at th e .0 1 le ve l( tw o- ta ile d) .

** *C

or re la tio n is si gn ifi ca nt

at th e .0 01

le ve l( tw o- ta ile d) .

906 Public Management Review

modelled as an antecedent rather than a moderator, because the effects of POF, behavioural integrity, mission comprehension ambiguity and self-efficacy on perceived message quality were not expected to depend on the employee’s position. Because post hoc tests (Dunett’s C and Scheffe, respectively) indicated that the variable means did not differ significantly between first-level managers and lower middle managers, these subgroups were grouped in further analyses.

Multivariate analysis

A latent variable model was developed to test the hypothesized relationships between the constructs of interest. The actual analysis followed a two-step approach (Anderson and Gerbing 1988) using IBM SPSS 21 Amos (IBM Corp. Released 2013b). In the first step, a confirmatory factor analysis (CFA) was used to assess the fit of the measurement model to the data. A multi-factor measurement model was preferred to an integrated factor model based on a number of single-factor models, because the latent variables were expected to be correlated with each other (Hair et al. 2009). In the second step, the relationships between the constructs were estimated.

Step 1. Psychometric properties of the measures: the measurement model A multi-factor measurement model that included all survey items was estimated using maximum likelihood estimation with bootstrapping (5,000 bootstrap samples). All variables were treated as latent constructs. We tested the measurement model for construct, convergent, discriminant and nomological validity. First, the construct validity of the included constructs was supported because the

loading of each factor was significantly different from zero and non-trivial (absolute standardized loadings >.60). The chi-square of the measurement model with all constructs was χ2260 = 1776.9 (p < .000). A significant chi-square could have been expected given the large sample size, because this particular index is sensitive to sample size; larger samples inflate the chi-square and decrease the likelihood of achieving good

Table 3: Mean study variables of hierarchical position subgroups (with standard deviations in parentheses)

Variables Non-

supervisor First-level manager

Lower middle manager

Higher middle manager

POF 4.82 (±1.13) 5.22 (±1.1) 5.27 (±1.12) 5.78 (±.73) Behavioural integrity 4.32 (±1.18) 4.67 (±1.27) 4.81 (±1.43) 5.87 (±.56) Self-efficacy 4.04 (±1.28) 4.85 (±1.25) 5.16 (±1.26) 6.18 (±.52) Mission comprehension ambiguity 5.34 (±1.43) 4.66 (±1.52) 4.90 (±1.48) 2.96 (±.81)

Desmidt: Employee mission statement perceptions 907

model fit (Hair, Black, and Babin 2010). However, the more sensitive fit statistics provided evidence of acceptable model fit (CFI = .94, GFI = .90, NFI = .93, RMSEA = .06) because they all meet the required thresholds (Hair, Black, and Babin 2010).1 In addition to being highly related to their respective constructs, all item factor loadings were significant, and the explained variance ranged from .47 to .77. The average variance extracted and the construct reliability of each construct exceeded .50 and .60, respectively. Second, following Fornell and Larcker’s (1981) suggestions, evidence of convergent validity was provided via the significant size of the completely standardized factor loadings ([.68, .88], average λ = .80). Third, all constructs are discriminant-valid. There are no signals of poor discriminant validity as the largest bivariate correlation – between POF and employee mission engagement –- of .58 is well below the .85 threshold (Kenny 2012). Multi-collinearity tests suggest that multi-collinearity is not an issue (a low-average bivariate correlation of .20, no measure shared more than 42% of its variance and no variance inflation factors exceeding 1.7), thereby providing further evidence of the constructs’ discriminant validity. Finally, the Fornell and Larcker (1981) discriminant validity test and a collapsed factor discriminant validity test further testify to the constructs’ discriminant validity. The square root of variance extracted for each construct was greater than the correlations between the given construct and any other construct (Fornell and Larcker 1981). In addition, the model fit was significantly worse when two constructs were collapsed, and this applied to all possible construct pairs (all chi-square tests were significant at the p < .001 level; collapsed factor models: average CFI = .87, average GFI = .82, average NFI = .87, average RMSEA = .09). Fourth, the fact that the correlations between the constructs are all positive except for mission comprehension ambiguity, as expected based on theory, indicates nomological validity.

Step 2. Relationships between the latent variables: the structural model The second step of the data analysis process examined the significance and strength of each of the hypothesized effects with maximum likelihood estimation using bootstrap- ping (5,000 bootstrap samples). The calculated model fit indexes suggest that the theoretical model (see Figure 1) captured the pattern of relationships found in the data: CFI = .92, GFI = .90, NFI = .91, RMSEA = .07 (Hair, Black, and Babin 2010). Given the large sample size, the fact that the maximum likelihood chi-square (χ2264 = 2,032.33, p < .000) was inconsistent with overall model fit was not considered to be especially troubling (Hair, Black, and Babin 2010; Wright 2007). The validity of the tested hypothesized model was further analysed in two additional

steps. First, the mediation effect of perceived message quality was measured by conducting a series of mediation tests using bootstrapping for a model that included both mediation and direct effects (Preacher and Hayes 2008). Perceived message quality fully mediated the relationship with employee mission engagement for behavioural

908 Public Management Review

integrity and self-efficacy. Perceived message quality partially mediated the relationship between POF and employee mission engagement (95% bias-corrected bootstrap con- fidence interval: indirect effect .05 to .10, p < .01, and direct effect .45 to .63, p < .001) and between mission comprehension ambiguity and employee mission engagement (95% bias-corrected bootstrap confidence interval: indirect effect −.09 to −.04, p < .001, and direct effect −.12 to −.02, p < .01). Subsequently, the hypothesized structural model was extended with a direct effect of POF and mission comprehension ambiguity, respectively, on employee mission engagement. Second, given the detected significant association between an employee’s organiza-

tion position and all latent variables, the effect of an employee’s organization position on the latent constructs was added to the hypothesized model. Organization position2

was modelled as an ordered categorical fixed exogenous variable and coded cumula- tively as two dummy variables: the ‘lower manager’ group, employees who had a position of first-level manager or higher (organization position >1), and the ‘higher middle manager’ group (organization position >2). The standardized direct effect of organization position was significant for all latent variables with the exception of perceived message quality and employee mission engagement. Figure 2 presents the parameter estimates for the final structural model as standar-

dized regression weights and the explained variance of the endogenous variables. The model fit indexes suggest a relatively good model fit: CFI = .93, GFI = .91,

NFI = .91, RMSEA = .06. As with the original model, only the maximum likelihood chi-square (χ2310 = 2016.84, p < .00) was inconsistent with overall model fit. Again, given the large sample size, this result was not considered to be especially troubling (Hair et al. 2009). The bootstrap bias-corrected two-tailed tests of the path coefficients of all hypothesized relationships were significant (p < .05), the relationships were non- trivial (absolute values > .05), and the predicted directions had the expected signs, providing additional support for the accuracy of the theoretical model. The structural model indicates that the antecedents of perceived message quality are

affected by a respondent’s organizational position. More specifically, respondents with managerial positions display higher levels of self-efficacy (β = .25, p < .001), POF (β = .12, p < .001) and behavioural integrity (β = .11, p < .001) than non-supervisors and deemed the mission statement less ambiguous (β = −.16, p < .001). However, not all managers were similar. Higher middle management had higher levels of self- efficacy (β = .12, p < .001), behavioural integrity (β = .10, p < .001) and POF (β = .06, p < .01) than first-level management and lower middle management and perceived the mission statement as less ambiguous (β = −.13, p < .001). With respect to the hypothesized antecedents of perceived message quality, the results

confirmed that mission comprehension ambiguity has a direct negative effect on perceived message quality (hypothesis 2, β = −.19, p < .001), whereas behavioural integrity (hypothesis 3, β = .31, p < .001), self-efficacy (hypothesis 4, β = .21, p < .001) and POF (hypothesis 5, β = .37, p < .001) had direct positive effects on perceived message quality. In addition, the

Desmidt: Employee mission statement perceptions 909

conducted mediation tests showed that POF, behavioural integrity, self-efficacy and mission comprehension ambiguity fully mediated the relationships between the constructed dummy variable ‘management’ (β = .16, p < .001) and ‘higher middle management’ (β = .10, p < .001) and perceived message quality. Aggregated, these variables explained 34% of the variance in perceived message quality. One of the main goals of this study was to examine the relationship between perceived

message quality and employee attitudes. The constructed structural model was able to explain nearly half of the variance in employee mission engagement (R2 = .47) and provided support for the hypothesis that perceived message quality has a positive, direct effect on employee mission engagement (β = .22, p < .001). However, although the data supported hypothesis 1, the degree of explained variance could not solely be attributed to perceived message quality, because the direct effect of POF (β = .56, p < .001) on employee mission engagement was stronger, whereas mission comprehension ambiguity had a direct negative effect (β = –.09, p < .01) on employee mission engagement. In addition to the direct effects, the conducted mediation tests showed that perceived message quality partially mediated the relationships between POF (β = .08, p < .001) and mission comprehension ambiguity (β = −.04, p = .001) and employee mission engagement.

Person- organization

fit

Behavioural integrity

Self-efficacy

Perceived message quality

Employee mission

engagement

Mission comprehension

ambiguity

*p < .01

** p < .001

Lower management

Higher middle management

R2 = .05

R2 = .09

R2 = .03

R2 = .02

R2 = .34 R 2 = .47

0.22**

0.37**

0.31**

0.21**

–0.19**

0.11**

0.12**

0.25**

–0.16**

–0.13**

0.10**

0.12**

0.56**

–0.09*

0.06*

Figure 2: Structural model

910 Public Management Review

Finally, additional analyses indicated that all the structural relationships were sig- nificant, regardless of employee age, gender, organization tenure, functional tenure and employment level or organization role.

DISCUSSION

Although public organizations often formulate and communicate a mission statement (Weiss and Piderit 1999), the question of whether mission statements are effective communication instruments has remained largely unanswered. Our results provide valuable information with respect to this issue, because they indicate that high levels of perceived message quality are positively related to employee mission engagement. Therefore, these results support previous claims that a salient mission can increase employees’ intrinsic motivation (Park and Rainey 2008). So, is it time to shelve the debate about the relevance of mission statements? Perhaps not yet. First, congruent with information processing theory, the study findings suggest that

individual perceptions of message quality are primed by a number of factors and cognitions (Cornelissen and Werner 2014). Consistent with studies indicating that the perceived attrac- tiveness of organizational goals and missions is driven by a combination of individual, job and organization characteristics (Wright and Pandey 2011), our findings indicated that contingen- cies related to the message (mission ambiguity), the sender (behavioural integrity) and the receiver (self-efficacy and POF) are related to perceived message quality. Moreover, the study findings indicate that not all employees find their organization’s mission statement attractive, self-explanatory or capable of eliminating goal ambiguity. As a result, the study results directly oppose the often implicit assumption that underlying management-initiated, top-down com- munication is uniformly perceived by all employees (Hill et al. 2012) and that the mere dissemination of a mission statement will help diminish employees’ cognitive dissonance about an organization’s essence (Cornelissen and Werner 2014). Second, POF displayed the strongest relationship to perceived message quality. Because

mission statements are developed to raise employee awareness of the importance of organizational outcomes, employees could be expected to respond more to mission statements that express goals and values within their existing zone of convictions (Paarlberg and Lavigna 2010). However, the detected direct relationship between POF and employee mission engagement could indicate that mission statements are preaching to the converted. It could be that selective distortion biases the recipient’s information processing processes, leading employees with an already favourable disposition towards the organization to find the mission statement more attractive (Munro, Lasane, and Leary 2010). With respect to this issue, Cameron (2009) argues that persuasive messages have generally three possible goals: response shaping (i.e. creating a response to a new stimulus when an attitude towards the stimulus did not yet exist), response reinforcing (i.e. strengthening an attitude or behaviour that is advocated by the persuader but already

Desmidt: Employee mission statement perceptions 911

enacted by the receiver) and response changing (i.e. the persuasive message is targeted at changing values, beliefs, attitudes, intentions or behaviour). Because it appears unlikely that employees will have no pre-existing attitudes regarding the organizational goals and values or that existing unfavourable attitudes will change after reading the mission statement, the impact of mission statements could be limited to response reinforcing. Third, the greater the hierarchical distance between employees and the top of the

organization, the less likely employees are to understand the mission statement, to be able to link its message to their jobs and personal values, and to deem the organization’s management team as acting in accordance with the mission statement. The study findings thus lend additional credibility to previous statements indicating that organizational understanding of mission statements among non-managerial employees is less than forthcoming (Vandijck, Desmidt, and Buelens 2007). The significantly better scores of managers on all of the study variables are most likely attributable to the fact that their organization positions provided access to more information about the organization’s mission and more opportunities to reduce ambiguity via two-way communication prior to the study. Moreover, because mission statements are inherently a collection of ambitious, abstract labels with a high level of interpretive leeway (Chun and Rainey 2005b), it is essential that employees generate a shared meaning for these labels. In the design of this study, in which employees were asked to read and evaluate a copy of the mission statement, respondents with low levels of knowledge were not provided the opportunity to engage in collective meaning construction processes with the aim of resolving interpretive leeway issues. However, one could argue that this research setting strongly mimics the majority of mission statement communication processes, in which employees at the lower end of the organizational hierarchy must rely on asymmetric, mediated, downward communication with limited interaction possibilities to gather information on the mission statement (e.g. external news releases, the organization’s website, internal newsletters, posters and leaflets), and efforts to reduce mission state- ment ambiguity are left to individual discretion. Consequently, the process of issuing a mission statement is insufficient to generate an organization-wide sense of mission, because ‘the inspirational power of public missions is not a given; [but] needs to be cultivated’ (Wright, Moynihan, and Pandey 2012, 212).

KEY CONTRIBUTIONS AND LIMITATIONS

Despite the omnipresence of mission statements within the public sector, this study is one of the first to focus on employee perceptions of an organization’s mission statement by adopting research methodologies from formative research. However, despite the study’s contributions, some limitations should be noted. For example, collecting data within one organization enabled a comparison of mission perceptions among the employees of the same organization but limits the generalizability of the study findings. Further research should

912 Public Management Review

concentrate on expanding the scope of this study and examine the extent to which the results are generalizable to other sectors, types of organizations and different employee groups. Second, a formative research design requires including a copy of the message in the questionnaire. Including a copy of the mission statement guarantees that all respondents will assess the properties of the same message but does not allow an assessment of top-of-mind perceptions of the organization’s mission or whether employees know that their organiza- tion has a mission statement. Third, the persuasive power of mission statements was measured using the concept of employee mission engagement. Future research should analyse how employee mission engagement is related to broader concepts, such as mission valence (Wright and Pandey 2011; Wright, Moynihan, and Pandey 2012) and mission attachment (Brown and Yoshioka 2003; Kim and Lee 2007). Fourth, each research model is a simplified representation of complex processes. Although the selected variables accounted for a large percentage of the variance in the dependent variables, other variables could be relevant to include in future research designs (e.g. job satisfaction, person–job fit and public sector motivation). Fifth, ‘both the dependent and focal explanatory variables are percep- tual measures derived from the same respondent’ using the same research instrument (Chang, van Witteloostuijn, and Eden 2010, 178). As a result, although various ex ante and ex post measures were taken to limit the occurrence of common method bias (see the section ‘Characteristics of the selected research design’ for a detailed discussion), the presence of CMV cannot be ruled out completely and should be taken into account when interpreting the study findings.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author.

NOTES 1 For models with N > 250 and between 12 and 30 observed variables: CFI ≥ .92, GFI ≥ .90, NFI ≥ .90,

RMSEA < .07. 2 1 = Non-supervisor, 2 = first-level manager or lower middle management (grouped), 3 = higher middle

management.

REFERENCES Anderson, J., and D. Gerbing. 1988. “Structural Equation Modeling in Practice: A Review and Recommended

Two-Step Approach.” Psychological Bulletin 103 (3): 411–423. doi:10.1037/0033-2909.103.3.411. Antonakis, J., S. Bendahan, P. Jacquart, and R. Lalive. 2014. “Causality and Endogeneity: Problems and

Solutions.” In The Oxford Handbook of Leadership and Organizations, edited by D. Day, 93–117. New York: Oxford University Press.

Bandura, A. 2012. “On the Functional Properties of Perceived Self-Efficacy Revisited.” Journal of Management 38 (1): 9–44. doi:10.1177/0149206311410606.

Desmidt: Employee mission statement perceptions 913

Bartkus, B., and M. Glassman. 2008. “Do Firms Practice What They Preach? the Relationship between Mission Statements and Stakeholder Management.” Journal of Business Ethics 83 (2): 207–216. doi:10.1007/ s10551-007-9612-0.

Brown, R. 1994. “Efficacy of the Indirect Approach for Estimating Structural Equation Models with Missing Data: A Comparison of Five Methods.” Structural Equation Modeling: A Multidisciplinary Journal 1 (4): 287– 316. doi:10.1080/10705519409539983.

Brown, W., and C. Yoshioka. 2003. “Mission Attachment and Satisfaction as Factors in Employee Retention.” Nonprofit Management & Leadership 14 (1): 5–18.

Cable, D., S. DeRue, and D. Scott. 2002. “The Convergent and Discriminant Validity of Subjective Fit Perceptions.” Journal of Applied Psychology 87 (5): 875–884. doi:10.1037/0021-9010.87.5.875.

Cameron, K. 2009. “A Practitioner’s Guide to Persuasion: An Overview of 15 Selected Persuasion Theories, Models and Frameworks.” Patient Education and Counseling 74: 309–317. doi:10.1016/j.pec.2008.12.003.

Chang, S., A. van Witteloostuijn, and L. Eden. 2010. “Common Method Variance in International Business Research.” Journal of International Business Studies 41 (2): 178–184.

Chun, H., and H. Rainey. 2005a. “Goal Ambiguity and Organizational Performance in U.S. Federal Agencies.” Journal of Public Administration Research and Theory 15 (4): 529–557. doi:10.1093/jopart/mui030.

Chun, H., and H. Rainey. 2005b. “Goal Ambiguity in U.S. Federal Agencies.” Journal of Public Administration Research and Theory 15 (1): 1–3. doi:10.1093/jopart/mui001.

Cornelissen, J., and M. Werner. 2014. “Putting Framing in Perspective: A Review of Framing and Frame Analysis across the Management and Organizational Literature.” The Academy of Management Annals 8 (1): 181–235. doi:10.1080/19416520.2014.875669.

Conway, J., and C. Lance. 2010. “What Reviewers Should Expect from Authors Regarding Common Method Bias in Organizational Research.” Journal of Business and Psychology 25 (3): 325–334.

Desmidt, S., A. Prinzie, and A. Decramer. 2011. “Looking for the Value of Mission Statements: A Meta-Analysis of 20 Years of Research.” Management Decision 49 (3): 468–483. doi:10.1108/00251741111120806.

Dillard, J., L. Shen, and R. Vail. 2007. “Does Perceived Message Effectiveness Cause Persuasion or Vice Versa? 17 Consistent Answers.” Human Communication Research 33 (4): 467–488. doi:10.1111/hcre.2007.33. issue-4.

Dillard, J., K. Weber, and R. Vail. 2007. “The Relationship between the Perceived and Actual Effectiveness of Persuasive Messages: A Meta-Analysis with Implications for Formative Campaign Research.” Journal of Communication 57 (4): 613–631. doi:10.1111/jcom.2007.57.issue-4.

Fairhurst, G., J. Jordan, and K. Neuwirth. 1997. “Why are We Here? Managing the Meaning of an Organizational Mission Statement.” Journal of Applied Communication Research 25 (4): 243–263. doi:10.1080/00909889709365480.

Favero, N., and J. B. Bullock. 2015. “How (Not) to Solve the Problem: An Evaluation of Scholarly Responses to Common Source Bias.” Journal of Public Administration Research and Theory 25 (1): 285–308. doi:10.1093/ jopart/muu020.

Fishbein, M., and J. Cappella. 2006. “The Role of Theory in Developing Effective Health Communications.” Journal of Communication 56 (S1): S1–S17. doi:10.1111/jcom.2006.56.issue-s1.

Fishbein, M., K. Hall-Jamieson, E. Zimmer, I. Von Haeften, and R. Nabi. 2002. “Avoiding the Boomerang: Testing the Relative Effectiveness of Antidrug Public Service Announcements before a National Campaign.” American Journal of Public Health 92 (2): 238–245. doi:10.2105/AJPH.92.2.238.

Fornell, C., and D. Larcker. 1981. “Evaluating Structural Equation Models with Unobservable Variables and Measurement Error.” Journal of Marketing Research 18 (1): 39–50.

Gioia, D., and K. Chittipeddi. 1991. “Sensemaking and Sensegiving in Strategic Change Initiation.” Strategic Management Journal 12 (6): 433–448. doi:10.1002/(ISSN)1097-0266.

914 Public Management Review

Hackley, C. 1998. “Mission Statements as Corporate Communications: The Consequences of Social Constructionism.” Corporate Communications: An International Journal 3 (3): 92–98. doi:10.1108/eb046557.

Hair, J., W. Black, and B. Babin. 2010. Multivariate Data Analysis. 7th ed. Upper Saddle River, NJ: Prentice Hall. Hair, J., W. Black, R. Tatham, R. Anderson, and B. Babin. 2009. Multivariate Data Analysis. Upper Saddle River,

NJ: Prentice Hall. Hill, S., M. Seo, J. Kang, and S. Taylor. 2012. “Building Employee Commitment to Change across

Organizational Levels: The Influence of Hierarchical Distance and Direct Managers’ Transformational Leadership.” Organization Science 23 (3): 758–777.

IBM Corp. Released. 2013a. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp. IBM Corp. Released. 2013b. IBM AMOS, Version 22.0. Armonk, NY: IBM Corp. Jakobsen, M., and R. Jensen. 2015. “Common Method Bias in Public Management Studies.” International Public

Management Journal 18 (1): 3–30. doi:10.1080/10967494.2014.997906. Jung, C. 2013. “Organizational Goal Ambiguity and Job Satisfaction in the Public Sector.” Journal of Public

Administration Research and Theory. doi:10.1093/jopart/mut020. Jung, C., and H. Rainey. 2011. “Organizational Goal Characteristics and Public Duty Motivation in U.S Federal

Agencies.” Review of Public Personnel Administration 31 (1): 28–47. doi:10.1177/0734371X10394404. Kenny, D. 2012. Multiple latent variable models: Confirmatory factor analysis. http://www.davidakenny.net/cm/

mfactor.htm Kim, S., and J. Lee. 2007. “Is Mission Attachment an Effective Management Tool for Employee Retention? An

Empirical Analysis of a Nonprofit Human Services Agency.” Review of Public Personnel Administration 27: 227–248.

MacKenzie, S., and P. Podsakoff. 2012. “Common Method Bias in Marketing: Causes, Mechanisms, and Procedural Remedies.” Journal of Retailing 88 (4): 542–555.

Maitlis, S., and T. Lawrence. 2007. “Triggers and Enablers of Sensegiving in Organizations.” Academy of Management Journal 50 (1): 57–84. doi:10.5465/AMJ.2007.24160971.

Meier, K., and L. O’Toole. 2013. “Subjective Organizational Performance and Measurement Error: Common Source Bias and Spurious Relationships.” Journal of Public Administration Research and Theory 23 (2): 429–456.

Moynihan, D., S. Pandey, and B. Wright. 2012. “Setting the Table: How Transformational Leadership Fosters Performance Information Use.” Journal of Public Administration Research and Theory 22 (1): 143–164. doi:10.1093/jopart/mur024.

Munro, G., T. Lasane, and S. Leary. 2010. “Political Partisan Prejudice: Selective Distortion and Weighting of Evaluative Categories in College Admissions Applications.” Journal of Applied Social Psychology 40 (9): 2434–2462. doi:10.1111/jasp.2010.40.issue-9.

Narayanan, V., L. Zane, and B. Kemmerer. 2011. “The Cognitive Perspective in Strategy: An Integrative Review.” Journal of Management 37 (1): 305–351.

Paarlberg, L., and B. Lavigna. 2010. “Transformational Leadership and Public Service Motivation: Driving Individual and Organizational Performance.” Public Administration Review 70 (5): 710–718. doi:10.1111/ puar.2010.70.issue-5.

Pandey, S., and J. Garnett. 2006. “Exploring Public Sector Communication Performance: Testing a Model and Drawing Implications.” Public Administration Review 66 (1): 37–51. doi:10.1111/puar.2006.66.issue-1.

Park, S., and H. Rainey. 2008. “Leadership and Public Service Motivation in U.S. Federal Agencies.” International Public Management Journal 11 (1): 109–142. doi:10.1080/10967490801887954.

Podsakoff, P., S. MacKenzie, and N. Podsakoff. 2012. “Sources of Method Bias in Social Science Research and Recommendations on How to Control it.” Annual Review of Psychology 63 (1): 539–569.

Podsakoff, P., S. MacKenzie, J. Lee, and N. Podsakoff. 2003. “Common Method Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies.” Journal of Applied Psychology 88 (5): 879–903. doi:10.1037/0021-9010.88.5.879.

Desmidt: Employee mission statement perceptions 915

Preacher, K., and A. Hayes. 2008. “Asymptotic and Resampling Strategies for Assessing and Comparing Indirect Effects in Multiple Mediator Models.” Behavior Research Methods 40: 879–891. doi:10.3758/ BRM.40.3.879.

Rainey, H. 1983. “Public Agencies and Private Firms: Incentive Structures, Goals, and Individual Roles.” Administration & Society 15: 207–242. doi:10.1177/009539978301500203.

Rainey, H., and P. Steinbauer. 1999. “Galloping Elephants: Developing Elements of a Theory of Effective Government Organizations.” Journal of Public Administration Research and Theory 9 (1): 1–32.

Ravasi, D., and M. Schultz. 2006. “Responding to Organizational Identity Threats: Exploring the Role of Organizational Culture.” Academy of Management Journal 49 (3): 433–458. doi:10.5465/AMJ.2006.21794663.

Shapiro, D. 2013. The power of words. Theme for the 74th Annual Meeting of the Academy of Management. http://aom. org/uploadedFiles/Meetings/annualmeeting/2014/theme.pdf

Simons, T., R. Friedman, L. Liu, and J. Parks. 2007. “Racial Differences in Sensitivity to Behavioral Integrity: Attitudinal Consequences, In-Group Effects, and “Trickle Down” among Black and Non-Black Employees.” Journal of Applied Psychology 92 (3): 650–665. doi:10.1037/0021-9010.92.3.650.

Smythe, W., D. Malloy, T. Hadjistavropoulos, R. Martin, and H. Bardutz. 2006. “An Analysis of the Ethical and Linguistic Content of Hospital Mission Statements.” Health Care Management Review 31: 92–98. doi:10.1097/00004010-200604000-00002.

Sonenshein, S., and U. Dholakia. 2012. “Explaining Employee Engagement with Strategic Change Implementation: A Meaning-Making Approach.” Organization Science 23 (1): 1–23. doi:10.1287/orsc.1110.0651.

Suh, T., B. Houston, S. Barney, and I. Kwon. 2011. “The Impact of Mission Fulfillment on the Internal Audience: Psychological Job Outcomes in a Services Setting.” Journal of Service Research 14 (1): 76–92. doi:10.1177/1094670510387915.

Urbany, J. 2005. “Inspiration and Cynicism in Values Statements.” Journal of Business Ethics 62 (2): 169–182. Updegraff, J., D. Sherman, F. Luyster, and T. Mann. 2007. “The Effects of Message Quality and Congruency on

Perceptions of Tailored Health Communications.” Journal of Experimental Social Psychology 43 (2): 249–257. doi:10.1016/j.jesp.2006.01.007.

Vandijck, D., S. Desmidt, and M. Buelens. 2007. “Relevance of Mission Statements in Flemish Not-For-Profit Healthcare Organizations.” Journal of Nursing Management 15 (2): 131–141. doi:10.1111/jnm.2007.15. issue-2.

Weiss, J., and S. Piderit. 1999. “The Value of Mission Statements in Public Agencies.” Journal of Public Administration Research and Theory 9 (2): 193–224. doi:10.1093/oxfordjournals.jpart.a024408.

Wright, B. 2007. “Public Service and Motivation: Does Mission Matter?” Public Administration Review 67 (1): 54–64. doi:10.1111/puar.2007.67.issue-1.

Wright, B., D. Moynihan, and S. Pandey. 2012. “Pulling the Levers: Transformational Leadership, Public Service Motivation, and Mission Valence.” Public Administration Review 72 (2): 206–215. doi:10.1111/ puar.2012.72.issue-2.

Wright, B., and S. Pandey. 2011. “Public Organizations and Mission Valence: When Does Mission Matter?” Administration & Society 43 (1): 22–44. doi:10.1177/0095399710386303.

916 Public Management Review

Appendix 1

Measurement items for all variables

Please read the included mission statement of [organization’s name]. Do you feel that the mission statement is.. .

Perceived 1. Impossible to enact – Feasible to enact message quality 2. Not worthy – Worthy

3. Not legitimate – Legitimate 4. Boring – Interesting 5. Boring – Exciting 6. Unimportant – Important 7. Common – Distinctive

Please read the [organization’s name] mission statement and indicate the extent to which you agree or disagree with the following statements.

Mission statement comprehension

1. This organization’s mission statement is clear to almost everyone who works here. (R)

Ambiguity 2. It is difficult to explain the goals listed in the mission statement to outsiders. 3. The mission statement contains clearly defined goals. (R) 4. It is difficult to explain the mission statement of this organization to a co-worker. 5. It is difficult to discuss organizational activities using the same words that are in

the mission statement. Behavioural integrity 1. The management of [organization’s name] practices what the mission statement

preaches. 2. The management of [organization’s name] does what the mission statement says

you need to do. 3. The management of [organization’s name] conducts themselves in accordance

with the mission statement. Self-efficacy 1. I have a clear understanding of how my job helps my work unit contribute to

[organization’s name] mission statement. 2. I have a clear understanding of how my work contributes to [organization’s name]

mission statement. 3. I can see how my work contributes to meeting the needs of external clients and

organizations as expressed in the mission statement. POF 1. The organization’s values and culture described in the mission statement provide

a good fit with the things that I value in life. 2. The things that I value in life are very similar to the things that my organization

values (according to the mission statement). 3. My personal values match the description of the organization’s values and culture

in the mission statement. Employee mission 1. I am motivated by the mission statement to do my work. engagement 2. I will work hard to ensure that (organization’s name) is successful in carrying out

the mission statement. 3. I carry out the mission statement when I do my work well.

Desmidt: Employee mission statement perceptions 917

Copyright of Public Management Review is the property of Routledge 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.