Why undertaking the change? A study on the recourse to goal framing in pursuing successful organizational changes

(1)

WHY UNDERTAKING THE CHANGE?

A STUDY ON THE RECOURSE TO GOAL

FRAMING IN PURSUING SUCCESSFUL

ORGANIZATIONAL CHANGES

MASTER THESIS

Author: Enrico Priora (s1910051)

Supervisor: Joris van der Voet

(2)

2

CHAPTER 1: INTRODUCTION

Introduction

The study of how organizational communication can help change makers to pursue successful changes in their organizations is receiving growing attention by Public Administration scholars (Beer & Nohria, 2000; Armenakis and Harris, 2002). Following this line of literature, the focus of this research will be the study of the relationship between change communication and employees’ reaction to it. Change communication is indeed recognized as crucial factor in stimulating workforce’s involvement in change attempts, involvement that in turn can prevent negative reaction by employees and prevent phenomena like resistance to change (Kotter, 1996).

Considering the broadness of the topic investigated, we chose to narrow down the study of this relationship by adopting as variables of our study the communication technique called goal framing, and its expected influence on employees’ reaction to change measured through the concept of readiness to change. In order to introduce the mentioned concepts, we start our analysis from a recent article by Andersen and Jakobsen (2017) that tried to shed light on how the recourse to strategic communication (which refers to the instrumental use of change communication in order to influence a determinate response by the recipients of the change message [Andersen & Jakobsen, 2017; Beer, 1980]) can help change agents to make bureaucrats more sympathetic toward a new policy. More specifically, the article found that presenting the introduction of a new policy by framing the change message in accordance to professionals’ norms and values is likely to make them sympathize with it.

Framing, which constitutes the starting point of our research, is there defined as “emphasizing one subset of considerations rather than others when describing an object, for example, emphasizing a certain aspect of a policy” (Andersen & Jakobsen, 2017, p. 57). The importance of framing is being increasingly recognized by public administration scholars: its value can be summarized in the words of Chong and Druckman (2007), who argued that through an appropriate use of framing, managers can cause “changes of opinion” into their audience (p.104), shifting this way potential negative reactions to change into attitudes of support for it. In this perspective, framing constitutes a powerful tool in the hands of change makers, who can use it to generate the expected reaction (e.g. support for the introduction of

(4)

4 new policy) into the organization’s workforce by conveying the change message in the proper way.

The study of specific communication techniques, among which framing, and employees’ reaction to the introduction of a new policy often intertwined in change management studies (Oreg, Vakola & Armenakis, 2011). A copious literature focused on the main causes leading to change failures suggests, indeed, that underestimating workforce’s involvement into organizational changes might result fatal for change agents (Armenakis, Harris, & Mossholder, 1993; Weiner, 2009; Fernandez & Rainey, 2006): when employees are not adequately involved, they might develop a negative attitude toward the change. Such a poor result, in turn, most of the times derives from insufficient attention paid by managers to internal communication (Barrett, 2002; Elving, 2005; Daly, Teague & Kitchen, 2003). According to this line of reasoning, an attentive use of strategic communication techniques by change agents emerges as crucial factor to successfully implement organizational changes, aimed to avoid negative behaviors from the workforce.

Building upon the previous mentioned literature, and particularly the Andersen and Jakobsen’s article, we try, in this project, to narrow down even more the study of framing and focusing on the cognitive technique known as goal framing, which will be adopted as independent variable in our theoretical model.

Goal framing refers to a communicative technique aimed to highlight positive and/or negative consequences of a choice (Chong & Druckman, 2007). Levin (2006) specifies that goal framing effects “occur when a … message has different appeal depending on whether it stresses the positive consequences of performing an act … or the negative consequences of not performing the act.” (p.413). This specific cognitive technique has been already studied in different fields (mostly among health, marketing and behavioral studies [Armenakis & Bedeian, 1999; Oreg et al., 2011]), but this concept is still quite recent to the change management literature. Andersen and Jacobsen’s article (2007) constitutes a positive exception, but the study of framing is still at its beginning. While their article focused on investigating how different framed messages can cause different reactions in communication recipients, our research will investigate potential reactions caused by listing the advantages and disadvantages of adopting (or not) the suggested behavior. Such investigation remains, as of today, an unexplored territory for change management scholars and this research is aimed to fulfill this gap, or at least to deepen the knowledge of this communication technique.

(5)

5 According to the classification made by Chong & Druckman (2007), the choice of highlighting positive or negative consequences of an act characterizes goal framing as positive goal framing or negative goal framing. More specifically, positive consequences are targets that the organization is actively trying to achieve. An example might be an organization wanting to implement a new IT system allowing officers to electronically sign documents from their terminals1_{. The change can be presented by managers highlighting its expected positive} achievements, like improving efficiency by archiving documents immediately into the organization’s system once they are signed. Negative consequences are, by contrary, occurrences that the organization is trying to avoid by changing, from which it is try to “escape” from. Following the previous example, the same change can be instead presented as aimed to reduce time and costs now linked to printing documents and sending them from one office to the other only to be signed in person. In the first case, a positive stimulus moves the organization (reaching a target); in the second case, the stimulus presents a negative nature (if the organization does not change, it might encounter adverse consequences) (Elving, 2005).

Both positive and negative goal framing, we argue, can have a positive effect on workforce’s reaction. Bases to argue a positive influence of positive goal framing might be found, for instance, in the literature regarding role of managers in organizations under change. One of their main task consists indeed of creating and communicating a powerful vision for their organization, vision usually defined as appealing picture of the future of the organization (Kotter, 1995). Presenting the positive consequences of changing can therefore result in positive attitudes of employees toward the new policy proposed by the management (Wanberg & Banas, 2000). On the other side, we expect negative goal framing to have a positive effect on readiness to change, too. Change makers should indeed not only be able to describe what it is desirable for the organization, but also what it is undesirable (Block & Keller, 1995; Mittal & Ross, 1998). Organizations change not only because they want to achieve specific targets: most of the times they do it because they want to avoid negative consequences that might derive from their inertia (Russ, 2008), e.g. suffering financial losses, or facing political pressure from the stakeholders because of poor performance (Nelson et al., 1997: Oreg et al., 2011).

Bandura & Pallak (1982) argue on this topic that in many cases moving away from potential negative future situations can be a stronger motivator to change than moving toward better

1_{Poste Italiane, the main Italian public operator in the field of mail delivery, recently adopted a similar IT system}

(extended to its clients) allowing its employees to electronically sign documents from their terminals. More information can be retrieved from the website: http://postecert.poste.it/firma/kit_retail.shtml

(6)

6 situations, due to the stronger fear that human beings experience from losing what they have compared to gaining what they only might have (also in Kahneman & Tversky, 1982). Authors like Nadler & Tashum (1989) and Fernandez and Rainey (2006) corroborated this hypothesis by pointing out the necessity by managers of instilling a sense of urgency to change into the workforce in order to make more vivid the need for change, while Spector (1989) argued that managers can sometimes even arrive to disseminate dissatisfaction regarding the organization’s current situation to stimulate employees reaction. In the words of Nadler and Tashum (1989), negative goal framing is expected to increase readiness to change because “the clear, present danger of organizational failure [i.e. the prospected negative consequences] creates the energy needed to make change happen” (p. 199).

Once we briefly introduced our independent variables, it is now time to devote our attention to our dependent variable: readiness to change. So far, it is been described as one the many measures used by public administration scholars to assess employees reaction to change. More technically, Armenakis (2007) defines it as an antecedent status of employees’ reactions to change, which can bring toward both commitment to change (a positive reaction) or change resistance (the negative counterpart). High levels of readiness are likely to positively prepare the terrain for change and induce positive attitudes among employees toward it; low levels of readiness usually lead instead to sabotative behaviors by employees (Beer, 1980). In other words, managers cannot expect support for change if its participants are not adequately activated, engaged and involved (Levin, 2006) before undertaking it. Failing to prepare employees might lead to lead to perceive change as threat by employees, or to not understand the reasons behind it, and causing at best a passive attitude, where participants are only spectators of change, or in the worst case scenario to an active opposition to change (Lewis, 2006).

Our thesis advances the hypothesis that goal framing might result to be a useful tool in the hands of change agents to higher the level of workforce’s readiness to change. We argue here that this technique, by specifying positive and negative consequences of an act, might help employees to perceive the purposes behind the organization’s decision to change (i.e. the appropriateness of the change), as well as the need for change that emerges by comparing the current state of the organization and what it wants to achieve or avoid. As we will have the chance to explain more in depth into the next section, these two factors, need for change and appropriateness, are at the basis of the multidimensional concept of readiness to change (Beer,

(7)

7 1980; King, 1984; Turner, 1982; Oreg et al., 2011): by leveraging them, we expect both positive and negative goal framing to lead to higher levels of readiness to change.

In order to study the relationship between goal framing and readiness to change, this research adopts an experimental design. Specifically, a vignettes survey will be administered to employees of a public organization in Italy (the Province of Asti, a local level of government). The questionnaire will introduce them a new working practice that the organization is considering to introduce as it would be done by the organization’s leadership, and will then ask them to rate their reactions. The main advantage of the experimental designs stands in the division of the population into control and treatment groups that allow us to present the same change message framed in three different ways. While the content of the message (the policy introduced) will not change across groups, it will be one time presented in a neutral form (only informing the workforce about the introduction of the change), one time one time negatively goal-framed (by presenting negative consequences of not-adopting teleworking), and one time positively goal-framed (by presenting positive consequences of adopting teleworking).

Problem definition and research question

This study aims to contribute to tackle one of the main concerns of change management scholars, how to prevent negative behaviors by employees when change is introduced (Fernandez and Rainey, 2006; Weiner, 2009; Pond, Armenakis, & Green, 1984), by adopting goal framing as independent variable and testing its potential positive impact on readiness to change.

The introduction of the concept of readiness to change is due to Armenakis et al. (1993), who defined it “the cognitive precursor to the behaviors of either resistance to, or support for, a change effort” (p.681). We move from this definition to argue that having the possibility to handle readiness to change could help change management practitioners to higher the success rate of their change attempts. Most of the literature on the topic, relying on numerous case studies of organizational change failures, focused on the causes leading to scarce involvement of employees in preparing the change, but only in few occasions tried to propose viable solutions to it (Weiner, 2009). Far from stating that this should be the main goal of these studies, we nevertheless decided to adopt a propositive perspective to advance the hypothesis that goal

(8)

8 framing can be the tool enabling change agents to solve this problem. Assessing readiness to change among the workforce and finding a viable way for managers to increase its levels is seen indeed as unavoidable step in every change attempt (Pond, Armenakis, & Green, 1984; Armenakis & Bedeian, 1999). Goal framing could be, in this perspective, the instrument to tackle the issue of resistance to change by preventing it (Weiner, 2009).

The main literature on strategic communication seems to suggest a potential positive role for goal framing in the process (King, 1984; Chong & Druckman, 2007). Due to its characteristics, goal framing, as example of strategic communication (King, 1984), might indeed match the requirements that the same literature considers crucial to effectively involve the workforce in change attempts: among the most important, providing information about why the organization decided to introduce a new policy, which reasons are leading the change, and why the current situation is not sustainable and needs to be changed (Russ, 2008; King, 1984).

Following these premises, our research will be driven by the following research question:

“What are the effects of positive and negative goal framing on readiness to change?”

Study relevance

This research draws from numerous studies focused on understanding the important role played by internal strategic communication in organizational changes, particularly in order to avoid change failures. The topic attracted a great number of scholars in the past decades (Lewis & Seibold, 1998) that starting from the necessity of understanding how to enhance workforce’s involvement in organizational change, devolved most of their efforts to analyze and explain the communication process. We argue, however, that there is still room to deepen the topic by focusing on a specific aspect of change communication: the recourse to goal framing to highlight positive or negative consequences that might follow the change, or the decision to not change (Andersen & Jakobsen, 2017; Beer, 1980), and so stimulate a reaction in the audience aimed to increase its levels of readiness to change.

While communication results to be a well-known subject to change management scholars, the same cannot be said about framing (Russ, 2008). Only recently, Andersen and Jakobsen (2017) have realized one of the few studies on the topic where they demonstrated that the way a policy

(9)

9 is framed and presented to professionals (aligning the change message to typical professionals’ norms) is “likely to make bureaucrats more sympathetic—or less negative—toward the policy” (p.1). Despite this recent article, however, most of the literature about framing found place in different study sectors: most of all behavioral and sociological studies (concerning individuals’ behavior in front of risky choices), marketing (where it has been applied to present more positively a product’s characteristics) and finally in the medical field in the attempt to increase breast self-examination attitude among women (on the various topics: Hannan & Freeman, 1984; Maheswaran, & Meyers-Levy, 1990; Meyerowitz, Chaiken, & Sherman, 1987). The research, therefore, aims to draw from the above-mentioned literature, to move toward a better understanding of the dynamics linking framing communication and readiness to change. Regarding the study’s social relevance, this dissertation aims to constitute a solid starting point for practitioners. The increased attention among scholars on the role of middle management (Lewis, 2006), due to their role of mediators between the top management taking strategic decisions and the employees called to implement them, suggests further research on which is the best way to communicate change and overcome (or even better prevent) negative employees’ reaction to it.

As anticipated by the previous section, the study will take the form of an experiment. The collaboration with the Italian local level of government “Province of Asti” allowed us to test our theoretical expectations into a practical context. The survey will consist of presenting the introduction of a concrete yet hypothetical change in the organization, namely teleworking as new working practice. Once the change will be introduced accordingly to its neutral, positive goal-framed and negative goal-framed articulation, employees will be asked to answer a short questionnaire aimed to assess their level of readiness to change. The value of this choice is twofold: on one hand, survey experiments allow researchers to measure the effects of a treatment (like the exposure to a goal-framed change message, in our case) on two (or more) randomly formed groups of employees, holding everything constant but the treatment itself. At the same time, this design is aimed to enrich the relevance of the study itself, under the consideration that such technique, even if not entirely new to the change management field (Grice & Montgomery, 2000), received nonetheless fewer attention compared to more widespread qualitative and Large-N approaches (Oreg et al., 2011; Devos, Buelens & Bouckenooghe 2007). Reason why the actual study aims to result innovative both in terms of

(10)

10 the studied relationship between goal framing and readiness to change, as well as for what concerns the methodology though which the study will be conducted.

Thesis structure

This thesis is divided into five chapters. The first chapter provided a general introduction of the topic investigated, its conceptualization into variables, and the research question that leads our research. The second chapter will present the theoretical framework on which our expectations are based: goal framing and readiness to change are here operationalized, and three hypotheses developed. Chapter 3 is instead dedicated to explaining in detail the research design and the data collection method, whose results are then analyzed in the fourth chapter. In the final chapter, the author will discuss the findings of the study, its academic, methodological and practical implications, together with its limitations, and provide suggestions to improve the present research.

(11)

11

CHAPTER 2: THEORETICAL FRAMEWORK

The previous chapter introduced readiness to change and goal framing, the two variables that constitute the focus of our study. The following chapter aims instead to describe them in more detail, and develop the theoretical expectations that will drive our data collection and analysis. We start by investigating the outcome variable, readiness to change, to understand its importance in change processes and discover its main components. We then turn to goal framing and its two articulations (positive and negative) trying to uncover the mechanisms linking them to readiness to change. The last section will summarize our theoretical assumptions and develop three hypotheses.

Employees readiness to change

We can start this chapter by locating the study of readiness to change in the broader area of studies focused on the implementation phase of changes (Beer & Nohria, 2000; Holt et al., 2007). Organizational managers introduce changes in their organizations to lead them toward specific goals (Holt et al., 2007). However, it is not uncommon that the path linking change’s ideation and implementation encounters problems (Weiner, 2009): the lack of support for change by the workforce has been recognized as the most important among them (Oreg et al., 2011). Spector (1989) and Weiner (2009) argues that in order to prevent behaviors of resistance to change, managers, even before implementing the change, should adequately prepare employees for it by presenting and explaining them the new policy and the reasons that motivate the change. In this perspective, an adequate preparation would drastically reduce the risk of negative reactions (Spector, 1989).

In this context, readiness to change emerges as critical factor to successfully pursuing change attempts (Weiner, 2009; Turner, 1982). Several scholars recall Lewin’s theory of change (1947) to explain its role. According to this author, organizations under change must go through three stages: unfreezing, consisting of preparing the organization to accept that change is necessary, moving/changing, where the change is introduced, and finally refreezing, where change is stabilized and internalized by organizations. In this context, readiness to change is part of the first step, helping managers to unfreeze the state of inertia proper of many organizations and their employees (King, 1984). Nadler & Tushman (1989) reinforce this view

(12)

12 pointing out that most organizations are change-resistant entities, and so do their employees, because changing is usually perceived as risky behavior (Mittal & Ross, 1998): if not imposed to them (by poor performance, the environment or by its stakeholders) they tend to maintain the same structure and processes over time, until changing becomes unavoidable (Katz & Kahn, 1978). It is therefore fundamental to get employees ready to change before entering the implementation phase, in order to avoid passive or negative reactions to it.

To better clarify the role of readiness to change, we can start from its definition by the relevant literature. Armenakis et al. define it as “the cognitive precursor to behaviors of either resistance to, or support for, a change effort” (2003, p.681). From a theoretical point of view, readiness to change has been studied for a long time together with the related concept of resistance to change (Beer & Walton, 1987; Turner, 1982). This definition by Armenakis, however, clarifies the distinction between them and particularly poses the accent on its function as precursor of both resistance and support for change, locating it a step before the two possible outcomes. Several authors, following Armenakis’ arguments, underlined the fundamental role played by readiness. Weiner synthetizes the main findings saying that “[w]hen organizational readiness for change is high, organizational members are more likely to initiate change, exert greater effort, exhibit greater persistence, and display more cooperative behavior” (2009, p.1). Implicit in the previous statement comes that when readiness is low, change attempts are instead likely to cause less support and more resistance to change efforts. While resistance to change refers then to a negative attitude by the workforce toward the change introduced by the organization, readiness to change stands in a previous stage of the process, where both positive and negative reactions (respectively resistance or support to change) can still be determined. The reason why we chose to focus on the precursor of support, or resistance, to change, instead of directly on them, stands in the observation that according to psychological studies on the topic (Hannan & Freeman, 1984), stimulating positive (or avoiding negative) feelings towards the change is a difficult reaction to convey, because it involves selling a determinate image of the change to employees (Green, 2002). Readiness, on the other hand, should be easier to achieve since it only requires to prepare the workforce to change (Wanberg & Banas, 2000): when trying to enhance readiness, managers are not therefore called to develop positive feelings towards the change, but only to set the right expectations about what advantages the change will bring to the organization, by adequately explaining the reasons behind it (Weiner, 2009).

(13)

13 Several scholars portrayed readiness to change as multidimensional concept involving different components. Weiner (2009), for instance, includes commitment to change and change efficacy among readiness components, where the first refers to a state of support for the change and the second to employees’ belief in the capability of the proposed change to tackle the organizational problems. Bandura & Pallack (1982) underline instead the role of self-efficacy, i.e. the extent to which employees believe they can carry out the change, while Spector (1989) highlights the importance of perceive that the change is useful to the organization and therefore is needed. The most comprehensive conceptualization is instead provided by Holt, Armenakis, Feild & Harris (2007) that listed five components of readiness (re-elaboration from Holt et al., 2007, p.251):

a) discrepancy: i.e. employees’ belief that the organization needs to change

b) appropriateness or organizational valence: i.e. employees’ belief that the proposed change is beneficial for the organization

c) management support: i.e. employees’ perception of top management’s commitment to change

d) change self-efficacy: i.e. employees’ belief on their capability of implementing the change

e) personal valence: i.e. employees’ belief that they will personally benefit from the change

The study of readiness in the change management literature usually followed an instrumental approach (Winer, Amic & Lee, 2008), usually directed to provide tools to change agents to solve the long-lasting problem of preventing resistance to change before its appearance. In this sense, by understanding how to enhance readiness to change from successful case studies, change management scholars aim to find a way of avoiding sabotative or passive behaviors by employees and increase the probability of change success (Russ, 2008). The mentioned approach helps us connecting the two variables of this study: readiness to change and, in a broad sense, change communication. The link between the two goes through the study of the strategies used by managers to enhance readiness. In the word of Armenakis et al. (2003), indeed, “[t]he primary mechanism for creating readiness for change among members of an organization is the message for change” (2003, p.1). Change communication plays therefore a central role in change processes.

(14)

14 Among these strategies, we find both positive approaches, used by manager to conquer and inspire their employees, and negative ones, aimed, instead, to stimulate sense of urgency to change (King, 1984). Weiner (2009) for instance points out that among the circumstances that favor increased readiness there are inspirational messages by leaders, aimed to communicate an appealing vision of the future. Armenakis et al. (2003) highlight instead the need for positive messages by change agents directed to enhance employees’ self-efficacy (as we saw one of the key components of readiness) to enhance their confidence in being able to manage the change. The literature makes clear, however, that positive communication is not the only way to increase readiness to change. On the contrary, the dissemination of negative information about the organization’s future seems to be, in certain contexts, even more powerful than positive messages (Lewis, 1999). In this perspective, communicating a sense of urgency and need to act is at the basis of many works in the literature: Bandura (1982) and Nadler & Tushman (1989) focus for instance on negative communication aimed to create intellectual pain (e.g. hypothesizing loss of jobs in the organization that might follow poor performance) to stimulate the workforce’s reaction, arriving sometimes to hypothesize the potential recur to individual threats to stimulate favorable reactions by employees. Similarly, Spector (1989) mentions the intentional dissemination of negative information by managers regarding the organization’s actual state (e.g. a high risk of cutbacks in light of below-expected results) with the intent of causing awareness of the undesired present situation.

The prevalence among successful case studies of communication strategies adopting a negative connotation seems to suggest a more effectivity of negative change messages. The argument finds also support in the literature. In their so-called prospect theory on decision making processes in risky choices, Kahneman & Tversky (1979) hypothesized for instance the presence of a negativity bias, according to which individuals tend to prefer avoiding a loss (negative prospect) than gaining a benefit (positive prospect). Following this line, negative formulated (or framed) messages could result more effective in shifting individuals’ position towards the suggested behavior. A second strong argument in favor of the prevalence of negative framed messages comes instead from the concept of negative potency. In its study on individuals’ formation of impressions about other individuals or objects, Vonk (1996) found that negative characteristics usually prevail over positive characteristic, even when they have comparable magnitude or are outnumbered by the positive ones. The observed effect is the result of individuals’ mental processes that lead us to evaluate quicker negative arguments than positive ones, usually because “negative [occurrences] are more often irreversible than those

(15)

15 [caused by] positive events” (Vonk, 1996, p.851). In other words, in case of dangerous situations people tend to react quickly and adapt their behavior, while the same does not happen when confronted with good occurrences. The negative potency stands therefore in the consideration that in front of equally perceived characteristics, or occurrences, negative ones are considered more salient by individuals than positive ones, resulting therefore in stronger reactions. The argument, however, is not exempt from critiques: Nadler & Tushman (1989) argue for instance that creating pain with the aim of preventing resistance to change can also result in counterproductive behaviors by employees, who develop feelings of closure and defensive reactions. Such reaction would in turn damage every change attempt (Spector, 1989). Moreover, we argue that the negative potency effect observed in other fields still has to be tested in the change management study area, a target that our research is aimed to achieve. Our study is therefore aimed to discern the possible outcomes on readiness to change that derive from applying the two opposite communicative approaches, the one presenting positive consequences of a change and the second underlining negative consequences of not acting. In this research, they have been named accordingly to the classification made by Levin (1998) and Chong & Druckman (2007) as positive and negative goal framing. The following subchapter will deepen these concepts and advance a theoretical model that involves the three variables.

Goal framing

We now turn our attention to the second variable of the research: namely goal framing. Before starting with a detailed description of goal framing, we briefly explain why the two variables will be studied together.

We previously said, in the introduction, that preventing resistance to change among the workforce is one of the most important challenge for today’s change management scholars. The way through which this problem can be overcome (or at least minimized) is recurring to the role of communication (Weiner, 2009). Managers, in this scenario, are called to convey the change message from the organization’s leadership to employees: given a change or policy that an organization want to introduce, there are multiple ways to introduce it to its staff members. Public administration scholars provide multiple examples of communication techniques aimed at this purpose: pointing out performance gaps between their own organization and similar ones (Spector, 1989), disseminating information (e.g. in form of reports) indicating lack of efficiency in the organization’s productive processes (Pond et al.,1984; Armenakis et al., 1993),

(16)

16 and making references to success models that can be imitated (Frumkin, 2004; Weick, & Quinn, 1999; Andersen and Jacobsen, 2016) are only few of them. This research chose one of the least studied (Levin, Gaeth, Schreiber & Lauriola, 2002), namely goal framing, to deepen the knowledge of the subject and measure its effect on employees’ readiness to change. Andersen and Jacobsen (2016) efficaciously explain that communication is an essential part of change attempts because “it surrounds every policy implemented by bureaucrats” (P.57): it has also been proved that communication can influence recipients’ attitude, which in turn can (and often do, as they specify) influence their behavior toward a policy or change. Chong and Druckman (2007) add a further piece to the puzzle by arguing that not only contents (and mediums) of communication are important, but the way change messages are framed plays an important role in the formation of attitudes too.

The major premise to study the impact of goal framing is that the way a policy is introduced and communicated is not neutral. How individuals interpret a change message, and consequently which attitude toward the change they develop, is the result of what the object of evaluation is and which arguments are used to present it (Chong & Druckman, 2007; Andersen and Jacobsen, 2016). Andersen and Jacobsen (2016) define indeed framing as “emphasizing one subset of considerations rather than others when describing an object” (p.59): by using this cognitive technique, change agents can decide what (and how) highlight when introducing a change, in order to strategically orientate employees’ attention on some attributes of the proposed change. What happens when a change message is framed is usually that “small … changes in the presentation of an issue or an event produce (sometimes large) changes of opinion” (Chong and Druckman, 2007, p. 104). In other words, change agents must take into account what they want to highlight of the new change and know that employees will react differently to the different ways they choose to describe it (Levin, Schneider, & Gaeth, 1998). It can therefore be extremely useful for managers understanding which technique would be more effective in conveying their change messages and the expected employees’ reaction to it, so that they can generate higher levels of readiness to change to prepare the workforce for the change’s implementation.

At least three different types of framing have been studied so far. Levin et al. (1998) summarized the literature on the topic and pointed out that they can be divided in risky choice framing, attribute framing and goal framing. All of them draw from the same assumption that “decision makers respond differently to different but objectively equivalent descriptions of the

(17)

17 same problem” (p.150): while the change message remains the same in its content, reactions change accordingly to its articulation. The first described, risky choice framing, entails measuring individuals’ response in front of choices that can lead to potential gains and losses for the decision maker. The main finding, which is also at the basis of Tversky and Kahneman’s (1981) prospect theory, is that individuals tend to take more risks to avoid losses than to realize gains (Mishra, Fiddick, & Smith, 2012). The second technique, attribute framing, consists of describing an object in terms of its attributes, deciding to highlight one characteristic or its complementary. A famous example is the piece of meat described alternatively as 75% lean or 25% fat, where one framing or the opposite does not change what meat is made of but it is able, nonetheless, to influence respondents’ choices. Contrarily to what seen for risky choice framing, we see positive framing prevailing in this case. Between two identical hams, the authors report indeed that the majority of people tended to choose the one described in positive terms (“75% lean”) over the same one but described with negative attributes (“35% fat”) (Levin et al., 1998). Ceteris paribus, the difference in choice was therefore due to the different framing here applied to the same object of evaluation.

These opposite findings concerning the first two framing effects, negative framing prevailing in risky choices and positive framing prevailing in attributes’ evaluation, make even more interesting the study of the third framing technique: goal framing. Contrarily to what just found, there are no conclusive evidences to argue, in this case, that one between positive and negative goal framing is expected to prevail. Before jumping to the conclusions, however, it is appropriate to start with its definition and the main mechanisms underlining it. Levin et al. (1998) describe goal framing as communicating a change highlighting its goal and consequences. Given a new policy to introduce into an organization, the change message can describe it in terms of positive consequences following the target’s implementation, or negative consequences of not implementing it. The articulation in positive and negative goal framing comes therefore from Levin et al’s (1998) observation that “positive frame focuses attention on the goal of obtaining the positive consequence (or gain), whereas the negative frame focuses attention on avoiding the negative consequence (or loss)” (p.168). It must be noted that the object of the change message, i.e. the new policy or change, remains the same in both formulations: what changes is only the way it is presented. The aim of goal framing is therefore persuasive and strategic (Nelson, Oxley, & Clawson, 1997): it aims to convince the message’s recipients that the proposed change will be beneficial for them and the organization. The organization’s leadership already decided what needs to be implemented: its goal is now to

(18)

18 convey positive reactions to its introduction, or least avoiding negative ones, by presenting benefits deriving from adopting the change or/and losses deriving from not adopting the change (Levin et al. 1998). Following the example already used in the introduction, a new IT system allowing employees to electronically sign documents from their terminal might be described as aimed to improve efficiency by archiving documents immediately into the organization’s system once they are signed (positively goal framed), or aimed to reduce time and costs now linked to printing documents and sending them from one office to the other only to be signed in person (negatively goal framed). The content of change is the same, the way it is framed points once to the advantages of changing and once to the disadvantages of not changing.

Previous findings on positive and negative approaches to change communication

As anticipated, there is no conclusive evidence that one of the two articulations could be more effective in conveying the hoped behavior. The main reason behind it is that goal framing is still a recent concept for change management scholars. Negative goal framing found nevertheless more confirmations of its effects (and prevalence) than positive goal framing so far.

We already mentioned that Kahneman & Tversky (1979) advanced the hypothesis of a negativity bias in human decision-making processes in front of risky choices, according to which people tend more likely to take a risk when consequences involve loosing something they already own than something that they might gain (also in Brewer & Kramer, 1986). Similarly, Rozin & Royzman (2001), describing the concept of negative potency, argue that the stronger effect of negative approaches might derive from “innate predispositions and experience … to give greater weight to negative entities (e.g., events, objects, personal traits)” (p.296): according to them, therefore, people tend to have greater reactions in front of negative occurrences due to the higher (and more vivid) perception of potential damage that might derive from them (also in Vonk, 1996). A further example from the health field is reported by Meyerowitz, Chaiken, & Sherman (1987), who found that stressing the decision to undertake a breast self-examination (BSE) in terms of negative consequences deriving by avoiding the behavior (e.g. not being able to take care of the illness since not prevented on time) resulted to be more effective than stressing the positive gain that could derive from undertaking the same exam (e.g. discovering to not suffer from the same disease). Several evidences, finally, have been found among decision-making studies, like Robberson & Rogers (1988) that focused on

(19)

19 encouraging physical exercises among people that don’t use to practice, or Newberry et al, (1993) on decisions to comply with tax duties when presented as potential losses caused by non-fulfillment instead of gains derived from compliant behaviors.

The above-mentioned studies seem to point out a stronger effect of negative goal framing in stimulating individuals to follow some pre-determined behaviors, but the literature is not consensual on it. Even among these studies, we find mixed results: it is not rare, indeed, that negative goal framing resulted in no effect (e.g. Lalor ad Hailey [1990] again on BSE-related decisions, or Reese et al. [1997] on psychological help to war veterans in the US). Moreover, in the threats and opportunities literature (where, according to Jackson & Dutton [1988] we define opportunities as positive expectations of gain and threats as negative expectations of loss following a specific behavior) we find several examples of counterproductive results deriving from use of threats and coercive methods. For instance, Dutton and Jackson (1997) contested the supposed effectivity of negative messages conveyed by threats, pointing out that even if threats, and more in general authoritative methods, usually succeed in forcing the hoped behavior on the short term, this happens at high costs by damaging the relationship between the change maker and the employees on the long term. Several studies on the subject confirmed their assumption, observing how threats can be counterproductive for organization’s purposes, leading to feelings of closure among employees that develop inactivity and negative reactions to change (Jackson & Dutton, 1988; Nadler & Tushman, 1989). Staw, Lance, Sandelands & Dutton (2010) named this phenomenon “threat-rigidity effect” (p.502), according to which individuals confronted with threatening situations tend to develop behavioral rigidity consisting in reduction of information processing (less information are acquired and processed by the individual) and lack of control of the situation: two elements usually leading to feelings of rejection of the new situation, in our case the change that the organization is aimed to introduce.

Finally, most of these authors also point out advantages that might derive from using opportunities as motivator to change, instead of threats. They argue, indeed, that highlighting the opportunities (i.e. the advantages) of a change is likely to make the choice of it more desirable for decision makers, resulting in higher rates of adoption of the suggested behavior (Green, 2002; Jackson & Dutton, 1988). The literature on this topic can be considered strictly related to goal framing: in both cases, managers indeed recur to the use of positive and negative expectations to push employees to adopt the wanted behavior (Pond et al.,1984).

(20)

20 Theoretical expectations

Despite the contrasting findings, we can nonetheless building upon the mentioned studies, which contributed to uncover some of the mechanisms of goal framing, and extend its study to the change management field to develop our theoretical expectations.

Relying on the description of readiness to change given by Holt et al. (2007) and reported in the previous section, our outcome variable resulted to be a multifaceted concept consisting of discrepancy, organizational valence, management support, self-efficacy, and personal valence. Basing our expectations on the theoretical framework, we argue that goal framing can influence at least two components of readiness: discrepancy and organizational valence. This expectation might be valid both for positive and negative goal framing, which, despite different mechanisms, are expected to cause similar effects on employees. Further effects of goal framing on the three remaining components are not to be excluded and they might result from our experiment: for the reasons listed below, we consider nonetheless more likely for goal framing to leverage the two mentioned readiness’s components.

Starting from negative goal framing, which received so far more attention in the literature, we argue that due to its characteristic this type of communication will result particularly effective in stimulating a sense of discrepancy and appropriateness of change. According to Armenakis et al. (1993), for instance, managers should highlight the need for change of an organization when a new policy is about to be introduced: the way they should do that is highlighting the difference between “the desired end-state and the present state” (p. 685). This implies that their message have to focus on the gap that separates the organization from the actual place (e.g. its actual performance) to where it should be (e.g. the desired performance): the larger the gap, the higher the sense of urgency conveyed and the stronger the stimulus to act to fill this gap (Fernandez and Rainey, 2006). The usefulness of recurring to comparisons is also highlighted by Gill (2002) according to whom the organizational change is “a journey from its current state to a desired future state” (p.311). The same belief of discrepancy can be stimulated through the recur to the dissemination of negative information about the organization (Armenakis et al., 2003; Bandura & Pallak, 1982; Nadler & Nadler, 1998) and the artificial creation of dissatisfaction toward the actual status (Fernandez and Rainey, 2006; Bandura & Pallak, 1982). All this ways, the authors argue, are aimed to make the audience aware that the organization must change in order to perform (Fernandez & Rainey, 2006) and that the proposed change is able to bridge the gap between the current state and the

(21)

21 desired one. It is aimed to unfreeze the workforce (Armenakis, & Harris, 2002) making it aware that the organization needs to change to survive (Russ, 2008), and results increased when fear-arousing beliefs are stimulated, and the negative consequences of inactivity are vividly suggested to the audience (Meyerowitz et al., 1987).

An increase of what we called organizational valence is also expected: organizational valence, or appropriateness, consists as said of employees’ belief that the proposed change is beneficial for the organization. The organization’s management might use the presentation of negative consequences with the purpose of making the change appear of service to the organization in order to solve its current problems. When a change is presented as solution of actual problems, is indeed likely to meet the favor of people expected to adopt the suggested behavior (Katz & Kahn, 1978). A negative use of goal framing it is therefore believed to leverage the mentioned components of readiness due to the explicit reference to negative consequences of inaction. The mentioned expectations can be summarized in the following hypotheses:

H1: negative goal framing is positively related to readiness to change H1a: negative goal framing is positively related to discrepancy

H1b: negative goal framing is positively related to organizational valence

The gap between the desired end-state and the present state, described before, is however not only typical of organizations affected by poor performance or in need to change. An organization may, indeed, perform very well and still want to change (Provan & Lemair, 2012), for instance in order to further improve its results or simply as reaction to external contingencies like the invention of a new technology (Frumkin, 2004). Even when organizations are underperforming, however, the management can choose to not point their attention to what they want to avoid (negative consequences) but on what they want to achieve: in this case, their intent of highlighting the discrepancy between current performance and desired one will be based on the advantages offered by the change, advantages that will be used here as term of comparison with the actual situation. In other words, while for negative goal framing discrepancy is represented by negative gaps (the negative occurrences that the organization might encounter if it does not change), here it is represented by positive gaps (what the organization might obtain by changing).

(22)

22 Similarly to the expected effect on discrepancy, recurring to positive goal framing might positively influence the perceived organizational valence to change. Too often, it is argued, employees adopt a negative, or passive, behavior in reaction to a new policy not because they oppose it but because they do not understand its potential for the organization (Russ, 2008). By highlighting positive consequences that might derive from change, managers can therefore uncover positive gaps (e.g. performance improvements) that the organization is aspiring to achieve (Levin et al., 2002). Higher levels of organizational valence might therefore derive from the managers’ attempt to make employees more aware of the advantages of changing, which in turn might result in higher levels of readiness. We can summarize this second block of hypotheses as follows:

H2: positive goal framing is positively related to readiness to change H2a: positive goal framing is positively related to discrepancy

H2b: positive goal framing is positively related to organizational valence

A third hypothesis finally completes our theoretical model. In addition to verify if negative and positive goal framing have positive influence on readiness to change, we are also interested in discovering which of the two framing effects will result to be the more powerful, conveying higher levels of readiness in the workforce. Due to the previously mentioned studies, however conducted in different fields than change management, we might expect stronger effects by negative goal framing, possibly due to the presence of negativity biases or the so-called negative potency. According to this last observation, we can develop the following third HP:

H3: negative goal framing is more positively related to readiness to change than positive goal framing

A further development of this research might also involve the measurement of the combined effect of the two techniques. Do positive and negative goal framing cause a stronger reaction when applied together than when applied individually? Or they cancel their respective effects out due to conflicting mechanisms? Unfortunately, a forth hypothesis of this type did not fit

(23)

23 this research due to the limited dimension of our case study population. We will nonetheless return to this topic in the conclusive section on suggestions for future researches.

(24)

24

CHAPTER 3: RESEARCH DESIGN

Chapter 3 starts by introducing the research design, followed by the explanation of the data collection method used in the research, and the operationalization of the study’s variables. The last two sections are dedicated to present the case study, and the study’s reliability and validity.

Research Design

The chosen research design for this thesis is an experimental study. Several factors contributed to this choice: firstly, topics involving the study of communication practices are fertile field for the application of experiments, used to measure the differences that small shades of the same message can cause in the audience. Even if not commonly employed in change management studies, experiments found great use among scholars of marketing and communication (Grice & Montgomery, 2000); as highlighted before, only recently Andersen and Jacobsen (2017) bridged the gap between the two fields in their study about framing effect and professionals’ reaction to the introduction of a new policy. Secondly, the present research adopts a theory-testing goal to verify if its theoretical expectations may find confirmation in the organizational reality: experiments are considered the gold-standard technique to pursue this goal (Shadish, Cook, & Campbell, 2002), being usually “set up to … establish and estimate a causal effect of practical importance” between the studied variables (Toshkov, 2016, Chapter 7, p.4). Thirdly, the random sampling distribution of participants ensures the similarity of samples among the three groups, approximating the creation of two counterfactual situations perfectly comparable to the control group. Fourthly, the splitting of participants across the three groups, each of them subjected to a different treatment by the researcher, allows an easy comparison of findings resulting from the comparison of the control group’s survey answers with the others’. Finally, experiments free the researcher from controlling for possible alternative variables that might influence the studied relationship by keeping everything constant among the groups but the treatment: therefore, there is no need to control for potential confounding factors (Toshkov, 2016).

(25)

25 Data Collection Method

The experiment took place by dividing the study population into one control and two treatment groups, each of these then exposed to a change message manipulated by the researcher. The study was conducted by administering online surveys through the software Qualtrics. An anonymous link to complete the survey was distributed by email to the workforce of a local level of government in Italy, and posted on the organization’s intranet, to invite all staff members to take part of the research:. Each employee was allowed by the system to take the survey only once. The survey’s population was automatically randomly distributed into control and treatment groups via Qualtrics.

The administered questionnaire consisted of three parts. The first part was dedicated to present an overview of the survey and the rules to complete it, explain the purpose of the study and provide the necessary context to facilitate employees’ understanding of technical terms used in it (e.g. the term “teleworking”). The second part asked the workforce to provide demographic background data for statistical analysis: gender, age, level of educational, organizational level, years of permanence in the organization. The questionnaire made clear that the survey would remain completely anonymous: therefore, no personal data (e.g. email address) allowing identifying specific employees were collected. The third part was finally used to present vignettes to the workforce: here, a realistic yet hypothetical change was presented as it would be communicated by the organization’s top management. This part, constituting the core of the experiment, varied between control and treatment groups depending on the goal framing technique applied to the change message: the change message was presented in a neutral form to the control group, negatively goal-framed to the first treatment group, positively goal-framed for the second treatment group. After being exposed to vignettes, employees were asked to assess their level of readiness to change by answering five questions aimed to measure its different dimensions. A sixth question, added in accordance with the hosting organization, was aimed to measure the predicted self-reported behaviors by employees in terms of their willingness to make use of the change (i.e. if they personally would recur to the practice of teleworking once introduced).

(26)

26 Operationalization Of Key Variables

This study involved the measurement of three variables: as independent variables, goal framing was articulated into positive goal framing and negative goal framing to reflect the expectations developed in the theoretical framework. As outcome variable, the study adopted the concept of readiness to change.

Readiness to change - operationalization

As anticipated by the previous chapter, readiness to change is considered by the change management literature a multi-dimensional variable (Armenakis et al., 1993). In order to measure it, this research makes use of the five items scale developed by Holt et. Al. (2007). The authors elaborated this scale by conducting a comprehensive review of readiness’ assessment methods previously adopted by other scholars in the field and elaborating their own items through a process of “item development, questionnaire administration, item reduction, [and] scale evaluation” (p.1), which, we argue, makes this scale suitable to reach an adequate level of measurement’s precision.

The five final components, resulted from their review, reflect the multidimensionality of the concept: here we find change self-efficacy, personal valence, discrepancy, organizational valence, and management support. Change self-efficacy refers to the degree employees believe they have the necessary skills to complete the tasks required to implement the change proposed by the organization, while personal valence assesses the perception by employees to personally benefit, in some form, from the introduction of the change. These first two components reflect a more personal perspective of readiness, related to employees’ self-perceived role in the change process. Discrepancy measures instead the perceived need for change, the extent employees believe there are good reasons to undertake the change by the organization. Organizational valence, slightly different from the previous component, focuses on the potential benefits for the organization to change, asking employees to assess if the change will result in improvements for the organization. Management support, finally, focuses on the perceived commitment by the top management to implement and support the change attempt. Reflecting this categorization, each readiness’s component was assessed through a separate question. In order to do this, employees were asked how much they agree with pre-formulated statements (e.g. "I think there are legitimate reasons for my organization to introduce this

(27)

27 change" for discrepancy) on a 10-point Likert scale, ranging from 1 (Fully disagree) to 10 (Fully agree).

Positive and negative goal framing - operationalization

The operationalization of the independent variables has been developed through the use of vignettes picturing the presentation of a realistic yet hypothetical change in the organization: the introduction of teleworking. We developed three different vignettes to reflect the distribution of the population in one control and two treatment groups.

The first scenario, distributed to the control group, presented the introduction of the change as it was announced the organization’s leadership in a neutral way, using the statement: “The Province of Asti is considering introducing teleworking in the next future”.

The following two vignettes, distributed to treatment group 1 and 2, presented the same content of the previous message, these times articulated accordingly to the goal framing technique. Scenario number 2 added to the above-mentioned change message the expected negative consequences deriving by not changing, using the expression: “in order to decrease the obligation by employees to be physically present every day at the workplace and avoid high offices’ renting and maintenance costs”. Almost symmetrically, the positively goal-framed change message presented the (expected) positive consequences deriving from undertaking the change by listing advantages like allowing “employees to organize with more autonomy their work and release financial resources now devoted to offices’ rent and maintenance”.

It must be noted that the two versions of the change message, besides presenting the same content (i.e. the introduction of teleworking), are also formulated as similar as possible in terms of number of words, as well as number and type of prospected consequences (every time in the number of two: one more focused on personal advantages/disadvantages for employees and one focused on advantages/disadvantages for the organization): we tried to avoid, this way, that any measured difference was due to the format rather than the framing of the message. The same attention was paid when translating the survey in Italian to facilitate its comprehension and enhance employees’ participation to the survey.

(28)

28 Case Selection

The study population is constituted by employees of a local level of government in Italy, namely the Province of Asti. The survey was distributed across a limited geographical area in the north of Italy: due to its dimension, structure and similarity of tasks, the Province of Asti is however very similar to almost all others homologous level of governments in Italy (Unione delle Provincie d’Italia [UPI], 2017). The whole workforce, constituted by 198 employees, was invited to take part of the survey via email address and by advertising the survey on the internal organization’s intranet. Each employee was allowed by the system to take the survey only once. The study population covered all roles present within this local level of government (from executive managers, to first and middle level managers and finally officers) as well as all the functional levels of the organization (e.g. directive offices, support departments, business-related departments).

In order to introduce an organizational change that employees could have easily recognized, the questionnaire presented the introduction by the organizational leadership of teleworking as new working practice that might become available to employees in the next future. Despite the hypothetical nature of the change, it represents a concrete potential development of the working practices at the Province of Asti: employees are indeed aware that, due to spending review’s measures introduced by the Italian central government, local levels of government are constantly required to implement new, more flexible, and potentially cheaper ways of working. Teleworking is also not entirely new at the Province of Asti, since employees have been granted to work from home in special circumstances (e.g. prolonged periods of illness). Finally, we believe the change suits the purpose of being presented according to positive and negative goal framing effects: together with the management of the local level of government, the researcher has indeed been able to identify at least two advantages given by the introduction of the new policy, together with two disadvantages that might follow the choice of not changing. The advantages and disadvantages have been then used to articulate the change message by goal framing it.

Reliability and validity of the study

A final observation on the method of assessment concerns its reliability and validity. Regarding the first, the researcher paid attention to fully explain the purpose of the survey

(29)

29 (without providing details concerning the treatments, which might invalidate the results) by explaining in clear terms what the questionnaire’s object was and what it was aimed to assess. Its formulation has also been kept as simple and short as possible, in order to allow employees to answer it without any difficulty. A section regarding possible remarks on the survey has also been provided: the absence of feedback or suggestions on difficulties found in filling in the survey might serve as indicator that the questionnaire has been fully understood by participants. We are therefore confident that our indicators provided a good level of reliability to the study.

A second concern was its validity. The choice of surveys as method of assessment guaranteed high internal validity of the assessment: one of the advantages of surveys, as said, consists indeed of allowing the researcher to measure the effects of the independent variable on the outcome variable, without the need to verify for any control variable affecting the relationship. Concerning istead its external validity, and so the generalization of the findings, as previously pointed out our case study, for demographic characteristics, dimension, departments and functions, is highly comparable to almost all others homologous level of governments in Italy (UPI, 2017), a characteristic that made our sample representative of the larger population of public employees working for same local levels of government. However, the composition of public organizations’ workforces might obviously vary across countries, and even inside each country depending on their level or functions. In our case, for instance, the average high age of employees might have played a role in influencing their readiness’ perception, while other public organizations might have an averagely younger workforce: we will reflect on this aspect, and its potential consequences, in the final chapter on study limitations and suggestions for future researches.

(30)

30

CHAPTER 4: DATA ANALYSIS

The following chapter is divided into three sections. The first one aims to provide an overview of the analytical strategy used to elaborate the data collected through the questionnaires. The second part analyzes the dataset in the attempt of providing a clear context for the results’ analysis. Finally, the last part analyzes the survey’s outcomes and verifies if the theoretical expectations found support in the collected data.

Analytical strategy

The study proceeds by first presenting the descriptive statistics concerning our case study and the variables investigated. These statistics provide the necessary context to proceed with more in-depth analysis of the investigated relationships. Secondly, we focus on readiness to change, our outcome variable; we previously saw that it presents five sub-dimensions: in order to see if these individually measured components can be summarized in a scorecard to obtain a single value of readiness, we will run a reliability analysis. Finally, the ANOVA and T-tests run on SPSS will help us to measure the groups’ results and verify potential significant differences among control and treatment groups due to the exposure to negative and positive goal framing.

Context

The first part of the chapter starts by analyzing our population’s characteristics (based on the demographic information collected through the questionnaires), the quality of the sampling distribution across groups and the descriptive statistics concerning the investigated variables, which help us provide context for the verification of the hypotheses.

The experiment was conducted on a total population of 198 employees at the Province of Asti, an Italian local level of Government. Out of 198 potential respondents, 74 employees completed the questionnaire: the response rate is consequently 37,37%. Before proceeding with the mentioned analyses of the dataset, the following chapter examines the quality of our sample and its homogeneity across groups. Our sample of 74 respondents has been randomly distributed via Qualtrics, the software used to administer the survey, across three groups: the random distribution assigned 23 participants to the control group (i.e. “group 1”), 26 to the first

Why undertaking the change? A study on the recourse to goal framing in pursuing successful organizational changes