Resistance and ReadinessDifferent measures or two extremes on the continuum of attitudes towards change?

(1)

Resistance and Readiness

Different measures or two extremes on the continuum of

attitudes towards change?

Master thesis, MscBA, specialization Change Management University of Groningen, Faculty of Economics and Business

Resistance and Readiness:

Different measures or two extremes on the continuum of

attitudes towards change?

Master thesis, MscBA, specialization Change Management University of Groningen, Faculty of Economics and Business

November 17, 2010 Oliver Hoffmann Studentnumber: 1621165 Winterberger Straße 27 51109 Köln Germany Tel.: +49 221 2786393 e-mail: Hoffmann.Olli@web.de Supervisor / University Cees Reezigt

Supervisor / field of study A. Lauterborn

RheinEnergie AG, Köln, Germany

Different measures or two extremes on the continuum of

(2)

ABSTRACT

(3)

TABLE OF CONTENTS

Introduction 5

The Company 7

Theory 9

Resistance and Readiness 9

Self-Efficacy 11

Participation 12

Organizational Commitment 14

Perceived Organizational Support 15

Methodology 16 Participants 16 Measures 17 Control Variables 17 Dependent Variables 18 Independent Variables 18 Data Analysis 19 Results 21 Correlation Analysis 21

Multiple Regression Analysis 25

Moderation of the Control Variables 30

Further Implications from the Multiple Regression Analysis 30

(4)

Discussion 36 Conclusion 42 Implications 43 Limitations 44 Further Research 45 References 46 Appendix 51

Appendix A: Questionnaire English 51

Appendix B: Questionnaire German 54

Appendix C: Descriptives of the Control Variables 58

Appendix D: Moderation of the Control Variables 59

Appendix E: Mediation Effects of the Antecedents 62

(5)

INTRODUCTION

For a change manager the goal of his task might seem to be rather simple. Within a project the management of the change is needed to communicate the advantages of the new ways of doing things and to identify and deal with opposing forces against these changes. Is change management really only dealing with opposing forces, usually named resistance to change? I do not think so. Sure, the change manager needs to identify the ones that do not want to go along with the changes, find out where the problems lie and create measures to counteract these forces. However, the change manager also needs to create an enthusiasm within the organization, a thrive to new dynamics and a strong commitment of all, who are affected by the change, to push into the same direction. So, can this be an easy task?

Over time the notion that resistance to change is not the only issue change managers need to deal with has become more and more famous. Resistance has a negative touch to it. It is defined as the opposing forces, the implications, which hinder change (Ansoff, 1990) or that what works against the change (Maurer, 1996). Sometimes authors have moved a bit away from the negative view and defined resistance as something helpful, because it shows weak points in the strategy, points to issues that have not been properly considered yet (Waddel and Sohal, 1998) and can also have positive influence on the change process (Ford, Ford and D’Amelio, 2008). For these reasons change management has, for a long time, only focused on the ones that are not committed to the change project. The question that arose was: “What is about the ones who welcome the change, the ones who want to contribute and are motivated to actively engage in the process?” In recent years authors such as Armenakis, Harris and Mossholder (1993) have started to think about these questions and developed another construct that should be on the agenda of every change agent: Readiness.

(6)

After readiness has become an accepted concept in the academic world and change managers started to shift their focus away from resistance and towards readiness new questions arise. Did managers find the counterpart to resistance? Is readiness the ying to the yang resistance? How do these two influence each other and what is their relationship? The answers to all of these questions have major implications for change managers within every organization. For the planning of future change efforts it is essential to know whether it is enough to focus only on one of the concepts because readiness and resistance are the two extremes on the continuum of change attitudes.

Many studies have analyzed the antecedents of the two concepts and based on their influences actions can be planned, such as involvement, communication or management support to counteract resisting forces or to strengthen driving forces. The trigger for the research at hand is the fact that the concepts and the antecedents have never been compared. Studies have always focused on either resistance or readiness, but some authors mention that a relationship should exist. Following the discussion in the introduction of the readiness article by Armenakis et al. (1993) both concepts are not seen as being clearly distinct from each other and creating readiness can be seen as a proactive step in dealing with resistance in the following change effort.

Armenakis et al. (1993), Oreg (2006) and Piderit (2000) complement each other in their view that the attitudes towards change need to be differentiated along three dimensions. One deals with the behavior (or the intended behavior) employees show when confronted with change, a second dimension focuses on the emotional aspects that are triggered by prospective changes and finally the third dimension deals with the cognitive processes of the ones affected by the change. Just the fact that resistance and readiness consist of these nearly similar dimensions raises the issue of comparability of the two.

(7)

influence both, in opposite directions. The aim is to either conceptualize the two concepts into one interrelated concept or to prove that they have to be seen as two distinct implications in change projects.

THE COMPANY

The RheinEnergie AG in Cologne, Germany is the fifth biggest energy supplier in Germany, with around 3000 employees. It is located in the Rhein-region, in the western part of Germany and delivers electricity, gas, water and heat to customers within the area.

In the recent past change management has become a big issue at the company. In Germany the market for electricity and gas changed dramatically when government liberalized the market, to open it to more competition and forced suppliers to install two different organizational entities, one for the supply and one for the distribution so that other suppliers can enter the market. This can be seen as the first major change for companies in the sector and it raised the issue of change management.

Nevertheless, from the perspective of the company, with regard to changes, this was only the start. Within the market for electricity and gas the dynamics are picking up pace at an incredible rate. The move towards greener energy production and the need of smart grids and information systems to meet governmental targets for reduction of green house gas emissions opens up opportunities and threats. New competitors are pushing on the market and established value chains need a complete overhaul.

The necessary movement away from being a sole energy supplier towards a service company puts new stresses and needed change projects on the agenda and as competitors are moving fast the changes need to be implemented quickly and in a highly efficient way. At the moment RheinEnergie is busy putting a change management structure in place to deal with the upcoming changes. The implications from the research at hand will be used for this change management structure to implement a quality management system.

(8)

those licenses are granted by region from regional governments, but in the future there is a great possibility that these licenses will be put out to tender on the European level. The final major change that occurs stems from the global discussion about the usage of water. People have started to decrease their usage, because of environmental and green issues. These threats change the market substantially and the department has to take measures to adapt to the future constraints.

The employees of the department are located at four major waterworks and at headquarters. All of them are affected by the upcoming changes and are included in the change management process. The diversity of the work force is great with regard to education and areas of activity.

(9)

THEORY

Many researchers have investigated the concepts of resistance and readiness. They have identified many antecedents that influence them both. The concept will be defined in the following section and the theoretical basis for the usage of the studied independent variables will be described. On the basis of past research hypothesis have been formulated, which will deepen the understanding of the relationship between the concept of readiness and the concept of resistance with regard to organizational change.

Resistance and Readiness

Resistance is a common research topic in today’s academic literature. The definitions within the literature are numerous, but complementary. Resistance is seen to be a factor in organizational change that hinders change (Ansoff, 1990) and works as persistence to avoid change (Maurer, 1996). However, resistance can also been seen as showing aspects of the change that have not been properly considered (Waddel & Sohal, 1998). Giangrecco and Pecci define resistance to change as “a form of dissent to a change process (or series of practices) that the individual considers unpleasant, disagreeable or inconvenient on the basis of personal or group evaluations” (2005: 1816f.). Piderit (2000) notes that in order to understand responses to organizational change in their full scope one has to study the concepts along multiple dimensions. Additional, Trader-leigh (2002) stated that resistance can occur on three different levels. Level one is where people question the idea, level two is when the resistance is deeper than the change and indicates other forces, which are at work and level three where there is deeply embedded resistance. Another distinction with regard to resistance is made by Oreg (2006). He defines resistance as a three-dimensional (negative) attitude towards change. For him resistance includes three types. Affective resistance deals with feelings about the change, cognitive resistance involves what one thinks about the change and behavioral resistance focuses on actions and intentions in response to the change. For the purpose of this research resistance is defined in line with Oreg as a “three-dimensional (negative) attitude towards change, which include affective, behavioral, and cognitive components” (2006: 76).

(10)

sub-concepts. The three groups that evolved are intentional, cognitive and emotional readiness for change and they come into play at different stages within the change effort (Armenakis et al., 1993; Bouckenooghe, Devos and Van den Brook, 2009; George and Jones, 2001; Oreg, 2006; Piderit, 2000; Szabla, 2007). Following the definitions of the before mentioned authors, intentional readiness is the “the extent to which employees are prepared to put their energy into the change process”, emotional readiness is “the affective reaction toward change” and cognitive readiness is “the beliefs and thoughts people hold about the change” (Bouckenooghe et al., 2009: 599). For this research readiness reflects beliefs, feelings and intentions regarding the extent to which changes are needed and perceptions of individual and organizational capacity to successfully enact those changes (Armenakis et al., 1993; Bouckenooghe et al., 2009).

From the analysis of the two concepts, both resistance and readiness are seen as three-dimensional concept, differentiated along the dimensions of behavior, emotions and cognitive processes. Two of the underlying dimensions of readiness and resistance (e.g. Armenakis et al., 1993 and Oreg, 2006) are highly comparable on their respective dimension. Cognitive readiness and cognitive resistance both regard the way people think about the changes and how this shapes their responses toward the change. Emotional readiness and affective resistance stem from the feelings one has about the change and therefore ground on the same psychological concept. With regard to intentional readiness and behavioral resistance comparison is more difficult as the authors describe them as focusing on different points in time. Nevertheless, the intentions that employees have about their behavior in change projects should be comparable to the actual behavior they show during the implementation of a change. To test the idea, that both concepts are interrelated and constitute the extremes on a continuum on attitudes towards change, the following hypothesis evolves.

H1: Readiness and resistance will show a strong negative correlation coefficient of - 1 on all three comparable dimensions.

(11)

purpose of this research four concepts that have been found to be antecedents of readiness and resistance are used. The four antecedents are self-efficacy, participation, perceived organizational support (POS) and organizational commitment (OC). In the following sections the antecedents and their relationship with readiness and resistance is described and hypotheses are formulated.

Self-Efficacy

Perceived self-efficacy concerns people’s beliefs in their capabilities to mobilize the motivation, cognitive resources and courses of action needed to exercise control over events in their lives (Wood and Bandura, 1989: 364). It is defined as the perception that an individual has of his own ability to perform or to cope with a future task or situation (Ashford, 1988) or as an overarching judgment of an individuals’ performance capability (Gist and Mitchell. 1992). Taylor (1983) raises the issue that it is important to note that self efficacy is not about the general confidence one has, but about the individuals’ belief about their ability to cope with a specific situation (as cited in Ashford, 1988). If an individual perceives him or herself as being able to adapt easily to changes he or she will be more receptive to organizational change efforts (Eby, Adams, Russell and Gaby, 2000).

(12)

Ellen, Bearden and Sharma (1991) studied the effect of self-efficacy on resistance in adopting new technologies. They found self-efficacy to have a strong negative impact on resistance. They conclude that people with high self-efficacy were less resistant to changing to the new program. Further support for the negative impact on resistance can be found in the study of Kim and Kankanhalli (2009). They concluded that self-efficacy reduces resistance to change indirectly through reducing switching costs.

However, some authors have shown empirical evidence that self efficacy has no influence on the change process (Van Dam, Oreg and Schyns, 2008; Eby et al., 2000)), but Oreg states that in his own opinion this stems from a too specific measure for role breadth self-efficacy (RBSE), which has been developed by Parker (1998), and the study of a merger where the employees did not have to make major positional changes. Armenakis et al. stated that “individuals will avoid activities believed to exceed their coping capabilities, but will undertake and perform those which they judge themselves to be capable of” (1993: 686). For this reason I will use self-efficacy as an antecedent for readiness and resistance. The relationship between self-efficacy and the two concepts of resistance and readiness is hypothesized as follows.

H2: Self-efficacy shows a positive correlation with all three dimensions of readiness, a negative correlation with all three dimensions of resistance and no significant difference exists between the correlations with readiness and resistance on the comparable dimensions.

Participation

(13)

conscious and intended effort by individuals at a higher level in an organization to provide visible extra role or role-expanding opportunities for individuals or groups at a lower level in the organization to have a greater voice in one or more areas of organizational performance” (1995: 402). For the purpose of this research participation is seen as being consulted about change related happenings (Beehr, Walsh and Taber, 1976), because this definition gives the greatest room for analysis of the relationships.

Many studies have found significant relationships between participation, resistance and readiness. Van Dam, Oreg and Schyns (2008) have shown that participation has a negative effect on resistance and Coyle-Shapiro stated that inviting employees to participate in a planned change increases the support for the change (1999). Lines (2004) studied specifically the influence of participation on resistance and discovered a significant negative correlation and Furst and Cable also found a significant negative relationship between their as consultation defined participation variable and resistance (2008). More support for the influence of participation on resistance comes from research done by Giangrecco and Peccei (2005), which produced a highly significant negative correlation between involvement in the change and resistance to change and Glew et al. state that participation is critical in reducing resistance to organizational change (1995).

(14)

H3: Participation shows a positive correlation with all three dimensions of readiness, a negative correlation with all three dimensions of resistance and no significant difference exists between the correlations with readiness and resistance on the comparable dimensions.

Organizational Commitment

(15)

employees are more likely to accept the change. Finally Vakola und Nikolaou (2005) also found a positive relationship of organizational commitment and positive attitudes towards change. Building on these findings the following hypothesis will be tested in this research.

H4: Organizational Commitment shows a positive correlation with all three dimensions of readiness, a negative correlation with all three dimensions of resistance and no significant difference exists between the correlations with readiness and resistance on the comparable dimensions.

Perceived Organizational Support

Perceived organizational support (POS) is a developed global belief of employees concerning the extent to which the organization values their contributions and cares about their well-being (Eisenberger, Huntington, Hutchison and Sowa, 1986). The authors explain further that “such perceived organizational support would depend on the same attributional processes, which people generally use to infer the commitment by others to social relationships” (1986: 501). The impact of POS on the readiness for change is documented in several research attempts. The perception of justification in change efforts is moderated by POS. Strong POS increases the perception that change is justified, therefore it lowers the impact on employees (Self, Armenakis and Schraeder, 2007), increases readiness and ultimately lowers resistance. Additional support can be found in the work of Eby et al. (2000), who found strong support that POS is positively related to readiness. Rafferty and Simmons (2006) proved that POS increases readiness for corporate transformation changes. The review of literature indicates that POS should have a positive impact on readiness for organizational change. Studies that link POS to resistance have not been done yet. Nevertheless, Burke (2003) found evidence that POS plays an important role in the development of job satisfaction during restructuring processes and has an effect on stressors. Following the academic research on POS the hypothesis that will be tested is:

(16)

and no significant difference exists between the correlations with readiness and resistance on the comparable dimensions.

Having defined the concepts of readiness and resistance, as well as the antecedents, which are used in the research, it is possible to test the comparability of readiness for change and resistance to change. The method of the analysis is outlined in the following section and is set up to test the hypothesis formulated from past research.

METHODOLOGY

The research consists of two interrelated steps. The first is a desk research, in which the concepts of readiness and resistance are defined and measures for the variables are identified. Furthermore, variables are identified that have been proven to have an influence on both concepts. They are needed to get a deeper understanding of the relationship between readiness and resistance. In the desk research the change literature regarding change was analyzed using electronic databases (Business Source Premier, Econlit, Academic Search Premier and databases on psychology); the articles were scanned on their fit with the purpose of the study, used to define the variables and to identify reliable measures, which are integrated in the questionnaire. Already existing measurement scales were used to ensure comparability with earlier studies and to research the relationship between the already defined concepts readiness for change and resistance to change.

In the second step the questionnaire was distributed among the employees of a department who face major changes in the future. During information events the employees were informed about the upcoming change process and given the questionnaires.

Participants

(17)

of the readiness and to assess the causes for resistance towards the change. The results are used in the process to define, plan and implement actions that will deal with the resistance and focus on the individual needs of the employees.

The whole department has a workforce of 149 employees, who are dispersed around several sub-departments that fulfill the necessary tasks to offer a good and efficient water supply for the area. Around 105 employees have participated in the information events. The information events were hold to inform the employees about the change process and the way in which the questionnaire will be used aside from the study. In total 101 employees completed the questionnaire, which results in a response rate of around 68%.

Within the group of the 101 respondents 86 are male and only 16 female. 2 employees are between 15-20, 6 are between 21-30, and 16 between 31-40. The biggest groups are those older than 40 years with 40 employees between 41-50 and 36 between 51-60. Just one employee was older than 60. Tenure and education have also been checked. The tables with all descriptives can be found in appendix B.

Measures

The aim of this research is to get a better understanding of the relationship between the established concepts of readiness and resistance. To ensure the comparability of the results with earlier studies on the concepts, measures have been used that have already been tested in past studies. For the purpose of empirical testing the following measurements are integrated in the questionnaire. All questions that have been used can be found in the questionnaire in the appendix A.

Control Variables. For control purposes and for deeper investigation of the subject four

variables were added to the questionnaire. With Age, Gender, Tenure and Education a sound basis for control can be achieved. Age, Tenure and Education were divided into groups (see questionnaire) to ensure anonymity of the respondents. The categorization produced ordinal scales.

Dependent Variables. Resistance will be measured using 15 items of the attitude towards

(18)

low factor loading and for cognitive resistance; Q14 has been dropped for the same reason. The Cronbach’s alpha values for the individual dimensions are .74, .72 and .77 respectively. The individual items will be measured using a seven-point scale ranging from “I absolutely do not agree” to “I absolutely agree”. Examples of the question include: “I was afraid of the change” (affective resistance), “I protested against the change” (behavioral resistance) and “I believe the change world make my job harder” (cognitive resistance).

“Readiness for change” is measured using nine items of the OCQ from Bouckenooghe, Devos and van den Broeck (2009). The items are divided into the three categories of Readiness: intentional; cognitive; emotional, with three items per category. From the analysis Q24 on cognitive readiness has been dropped, because of a too low factor loading. The individual items will be measured using a seven-point scale ranging from “I absolutely do not agree” to “I absolutely agree”. The factors have alphas of .66; .66; .82 respectively. Examples of the items are: “I want to devote myself to the process of change” (intentional), “Plans for future improvements will not come to much” (cognitive) and “I find the change refreshing” (emotional).

Independent Variables. Perceived Organizational Support (POS) is measured by using the 12

items of the short version of the survey developed by Eisenberger et al. (1986). The scale that was retained of the short version has an alpha of .92. The items are measured using a seven-point scale ranging from “I absolutely do not agree” to “I absolutely agree”. Examples are “The organization values my contribution to its well-being” and “The organization cares about my general satisfaction at work”.

Participation is measured using the three items of the non-participation scale from Beehr, Walsh and Taber (1976). The scale has an alpha of .51. The individual items will be measured using a seven-point scale ranging from “I absolutely do not agree” to “I absolutely agree”. An example of the items is “I am usually told about important things that are happening at RheinEnergie”. The scale produced an alpha of only 0.51. This is lower than the acceptable result for the usage of the scale. I have included the scale in the further research, but readers of this paper need to keep the low alpha value in mind when interpreting the results.

(19)

scale ranging from “I absolutely do not agree” to “I absolutely agree”. Examples of the items are “Wherever the restructuring takes me. I’m sure I can handle it” and “I have reason to believe I may not be able to perform well in my job situation following the restructuring”.

Organizational commitment is measured using 14 items from a scale created by Allen and Meyer (1990). The authors divided OC in three factors and out of each factor the five items with the highest factor loadings on their respective factor are used for the research at hand, for affective commitment only four items have been used. The three components of Organizational commitment, affective, continuance and normative commitment, are measured with four, five and five items respectively. The items have been phrased to reflect the specific situation that is to be studied. The individual items will be measured using a seven-point scale ranging from “I absolutely do not agree” to “I absolutely agree”. Examples of the items are “I do not feel a strong sense of belonging to my organization” or “I do not feel emotionally attached to this organization”. For the purpose of this study organizational commitment is studied as one concept, so that the items for all three components are combined. Four items had to be dropped, because they loaded too high on another factor. These items were Q41, Q43, Q51 and Q54. The 10 item scale has an alpha of .83.

Data Analysis

(20)

(21)

RESULTS

The following section presents the results of the statistical analysis. The outline of this part follows the structure as described in the methodology section. All relevant results are presented and further analyses can be found in the appendices.

Correlation analysis

Table 1 shows the correlation matrix and the descriptive statistics of the variables under investigation. The alpha values for each scale are given on the diagonal.

The means for the measures of readiness and resistance show that within the studied sample readiness is high and resistance seems to be rather low. This is true for all three dimensions. For perceived organizational support, participation and organizational commitment the means are around the middle of the scale, but still show more positive results. Only self-efficacy has a rather high mean of 5.73 and a lower standard deviation compared to the other measures.

Table 1 Correlation Matrix Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 1. Intentional Readiness 5.14 1.01 (,66) 2. Emotional Readiness 4.47 1.36 ,480** (,82) 3. Cognitive Readiness 4.78 1.35 ,391** ,466** (,66) 4. Behavioral Resistance 2.33 1.30 -,344** -,433** -,462** (,74) 5. Emotional Resistance 2.76 1.16 -,381** -,596** -,418** ,728** (,77) 6. Cognitive Resistance 3.44 1.27 -,431** -,785** -,532** ,469** ,595** (,72) 7. Organizational Commitment 4.73 1.22 (,196) ,040 ,107 ,099 ,120 -,075 (,83) 8. Perceived Organizational Support 4.37 1.24 ,261* ,347** ,360** -,140 -,234* -,382** ,475** (,92) 9. Participation 4.85 1.22 ,303* (,193) ,342** -,148 -,158 -,167 ,274** ,625** (,51) 10. Self Efficacy 5.73 1.03 ,329** ,288** ,228* -,365** -,451** -,334** -,075 ,236* ,305** (,69) 11. Age 4.04 .989 ,038 ,237* -,013 -,041 -,131 -,238 ,130 -,223* ,152 ,017 12. Tenure 4.42 1.08 ,031 ,160 ,008 -,126 -,154 -,236* ,151 ,126 ,220* ,009 ,554** 13. Education 2.47 1.49 ,226* ,165 ,210* -,226* -,183 -,186 -,158 ,029 -,011 ,239* -,077 -,331** ** Correlation is significant at the 0.01 level (2-tailed).

* Correlation is significant at the 0.05 level (2-tailed). () Correlation is significant at the 0.1 level (2-tailed). Cronbach’s alpha appears on the diagonal

(22)

dimensions of resistance are even higher. Significant correlations (p < 0.001) can be found between .469 and .728, which was again expected, because Oreg (2006) defines these three dimensions as reflecting the dimensions of resistance to change.

For the analysis of the first hypothesis the correlation matrix produces no supporting results. The correlation between intentional readiness and behavioral resistance is - .344, which is far from the in H1 expected outcome of - 1. On the emotional dimension H1 cannot be supported either. The correlation between emotional readiness and emotional resistance is

- .596, which is the highest correlation with regard to H1. Nevertheless, the correlation is too low to support it. The same can be said about the correlation between cognitive readiness and cognitive resistance, which results in - .532. All of these correlations are significant at the 0.01 level.

Table 2

Self-Efficacy Correlations

Readiness Independent Resistance Steiger’s Z

Intentional .329** Self-Efficacy -.365** Behavioral 4.17 (p< .01)

Emotional .288** Self-Efficacy -.451** Emotional 4.27 (p< .01)

Cognitive .228** Self-Efficacy -.334** Cognitive 3.24 (p< .01)

** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed)

(23)

Table 3

Participation Correlations

Intentional .303** Participation -.148 Behavioral 2.62 (p< .01)

Emotional .193 Participation -.158 Emotional 1.91 (p> .05)

Cognitive .342** Participation -.167 Cognitive 2.86 (p< .01)

For the relationship between participation and the dimensions of readiness and resistance, as stated in hypothesis 3, a somewhat similar picture evolves. With a Cronbach’s’ alpha of 0.51 the reader has to be aware of the implication that the scale is not as reliable as needed for a sound interpretation of the results and keep this in mind when drawing conclusions on the results of the study with regard to this scale. Surprisingly participation shows no significant correlations with either dimension of resistance. On the intentional and behavioral dimension the correlations are for readiness .303 (p < 0.01) and for resistance - .148 (p > 0.1). On the emotional dimension the distance between the absolute numbers of the correlations is very small, but they are not the same. Furthermore, they differ in their significance. For readiness the correlation is .193 (p > 0.05) and with regard to resistance the outcome is - .158 (p > 0.1). On the cognitive dimension the absolute coefficients are again far away from being similar. Readiness correlates with participation with a degree of .342 (p < 0.01) and resistance with - .167 (p > 0.1). When testing the comparable correlations on their significant differences Steiger’s Z shows the following results. On the behavioral dimension (z = 2.62) a significant (p < 0.01) difference is found, the same holds for the cognitive dimension (z = 2.86; p < 0.01). On the emotional dimension no significant difference can be found (z = 1.91; p > 0.05). Participation shows a similar influence on emotional readiness as it does on emotional resistance. Nevertheless, both correlations are not significant and can therefore not be used to support hypothesis 3, so that H3 is rejected.

Table 4 POS Correlations

Intentional .261* POS -.140 Behavioral 2.32 (p< .05)

Emotional .347** POS -.234* Emotional 3.20 (p< .01)

Cognitive .360** POS -.382** Cognitive 4.23 (p< .01)

(24)

Hypotheses 4 proposed that POS shows no significant difference between the correlation coefficients for readiness and resistance on each of the three dimensions. For this hypothesis very little support can be found in the results. The correlation of POS with intentional readiness is .261 (p < .05) and with behavioral resistance the outcome is - .140 (p > .1). On the emotional dimension the correlation for readiness results in a coefficient of .347 (p < .01) and for resistance of - .234 (p < .05). Correlations between POS and the cognitive dimensions show a coefficient for readiness of .360 (p < .01) and for resistance of - .382 (p < .01). These results give little support for H4, because the hypothesized influenced can be found on the emotional and the cognitive dimension. Similarity between the absolute values of the coefficients cannot be found as the results of Steiger’s Z show that on the behavioral (z = 2.32; p < .05), the emotional (z = 3.20; p < .01) and the cognitive (z = 4.23; p < .01) dimension significant differences exist. Therefore, H4 is rejected.

Table 5 OC Correlations

Intentional .196 OC .099 Behavioral 0.58 (p> .05)

Emotional .040 OC .120 Emotional -0.43 (p> .05)

Cognitive .107 OC -.075 Cognitive -1.08 (p> .05)

(25)

The first part of the analysis reveals that readiness and resistance do influence each other, but they are not two extremes on a continuum. Furthermore, their antecedents, which have significant correlation coefficients, show significant differences in their impact on either readiness or resistance.

Multiple Regression Analysis

The second step in the analysis of the hypotheses was to perform a multiple regression analysis. Assuming a model that includes all four independent variables to predict the dependent variables of readiness and resistance on their comparable dimensions, the regression should reveal beta’s, which are similar in value but opposite in sign, similarity between the degrees of variance explained within the dependent variable and the predictive power of the models would be similar, so that analysis results in a comparable F-Value. Furthermore, by calculating a predicted outcome for each case on the basis of the two comparable models the following test using Steiger’s Z with the correlation coefficients of the two predicted and the real variable will show whether significant differences exist between models. Tables 6, 7, 9, 10, 12 and 13 show the results for the multiple regression analysis for readiness and resistance on each dimension.

Table 6

Intentional Readiness: Multiple Regression Analysis

Intentional Readiness B beta R² F

Constant 1,938 0,191 5,025

Self Efficacy ,303 ,282**

Participation ,167 0,199

POS -,015 -0,019

Organizational Commitment ,147 0,178

(26)

as they show strong positive and significant correlations with intentional readiness. Even more surprising is the fact, that POS shows a negative beta of - .019, however, this value is not significant.

Table 7

Behavioral Resistance: Multiple Regression Analysis

Behavioral Resistance B beta R² F

Constant 4,838 0,147 3,503

Self Efficacy -,446 -,320**

Participation -,024 -,022

POS -,121 -,116

Organizational Commitment ,150 -,143

The model for the multiple regression analysis on behavioral resistance is similar to the one used for intentional readiness. Self-efficacy, participation, POS and OC together explain 14.7% of the variance in behavioral resistance and the F-Value of the model is 3.503, which is rather low. As for intentional readiness self-efficacy is the only significant contributor to the model with a beta of - .320. Participation, POS and OC result in low, negative and non-significant correlations. These directions of the influences are expected as they align with the outcomes of the correlation analysis.

I have used the unstandardized coefficients of the two models to compute two predicted variables for intentional readiness. Entering the resulting two variables together with the variable for intentional readiness into a bi-variate correlation analysis gives the matrix as shown in Table 8.

Table 8

Correlation Matrix of predicted Intentional Readiness

1 2 3

1 Intentional Readiness 1 ,437** -,291**

2 IR_Model1 ,437** 1 -,665**

3 IR_Model2 -,291** -,665** 1

(27)

Table 9

Emotional Readiness: Multiple Regression Analysis

Emotional Readiness B beta R² F

Constant 2,096 0,169 4,333

Self Efficacy ,287 0,203

POS ,459 ,430**

Participation -,168 -,152

Organizational Commitment -,085 -,078

The model that should explain the emotional dimension of readiness consists of the following independent variables: self-efficacy, POS, participation and Organizational Commitment. The four independent variables account for 16.9% of the variance within emotional readiness and have only a small predictive power, as shown in a F-Value of 4.333. Solely POS is a significant contributor to the model with a beta of .430. Rather unexpected is the finding that self-efficacy and participation do not have significant beta’s and that participation even shows a negative influence on the model. These findings are contrary to the correlations analysis where all three variables showed positive significant relationships with emotional readiness.

Table 10

Emotional Resistance: Multiple Regression Analysis

Emotional Resistance B beta R² F

Constant 5,399 0,254 7,16

Self Efficacy -,499 (-),423**

POS -,243 (-),273*

(28)

Again, the unstandardized coefficients of the two models are used to compute two predicted variables for emotional readiness. Entering the resulting two variables together with the variable for emotional readiness into a bi-variate correlation analysis gives the matrix as shown in Table 11.

Table 11

Correlation Matrix of predicted Emotional Readiness

1 2 3

1 Emotional Readiness 1 ,412** -,338**

2 ER_Model1 ,412** 1 -,821**

3 ER_Model2 -,338** -,821** 1

The comparison of the multiple regression analysis using the same models for readiness and resistance on the emotional dimension yields no support for either one of the hypotheses. Calculating Steiger’s Z using the correlation coefficients as stated in Table 11 shows that the models are significantly different with, Z = 3.88 and p < 0.01.

Table 12

Cognitive Readiness: Multiple Regression Analysis

Cognitive Readiness B beta R² F

Constant 1,571 0,181 4,651

POS ,302 ,274*

Self Efficacy ,216 0,149

Organizational Commitment -,039 -,035

(29)

Table 13

Cognitive Resistance: Multiple Regression Analysis

Cognitive Resistance B beta R² F

Constant 6,270 0,23 6,355

POS -,474 -,476**

Self Efficacy -,372 -,282**

The standard model is able to explain 23% of the variance when used in a multiple regression analysis on cognitive resistance. The predictive power is moderate with a F-Value of 6.355. As expected from the correlation analysis only POS and self-efficacy show significant contributions to the model and yield beta’s of - .476 and - .282 respectively. Participation and OC do not contribute to the regression on a significant level.

Finally, Table 14 shows the correlation coefficients of the two predicted variables that are calculated using the unstandardized coefficients from both models with the variable for cognitive readiness.

Table 14

Correlation Matrix of predicted Cognitive Readiness

1 2 3

1 Emotional Readiness 1 ,426** -,347**

2 ER_Model1 ,426** 1 -,816**

3 ER_Model2 -,347** -,816** 1

The in hypothesis1 proposed comparability of readiness and resistance on the cognitive level cannot be supported, as the same compilation of the model yields different regression results. The similar values proposed for the independent variables with regard to their contributions to the model cannot be supported either. None of the beta values are comparable as all independent variables have different results on the two distinct multiple regressions. The models are significantly different as Steiger’s Z results in a value of z = 4.00, which implies a significant difference between the models with p < 0.01.

(30)

continuum as their antecedents show significant differences in the influence on either one of them on their comparable dimensions.

Moderation of the control variables

While gender, age, tenure and education have been included in the questionnaire as means to control for differences within the sample, no substantial changes in the correlations, with regard to the hypotheses could be found. Gender was excluded from the study as only 16 women answered the questionnaire, and the size of this subgroup is too small for further analysis. For the remaining control variables moderating effects regarding the correlations between independent and dependent variables, as well as for the comparison between the dimension of readiness and resistance can be found, but not to the extent that they would support one of the hypotheses under investigation. None of the correlations of the different subgroups show a correlation between readiness and resistance of -1 or have close absolute correlations between the independent and dependent variables that cannot be found in the overall sample. Further investigation of the moderating effects in readiness and resistance could be interesting but would exceed the scope of this research, as the purpose is to investigate the comparability of readiness and resistance and not the concepts and their antecedents individually. For this reason, further elaboration on the mentioned differences is not included in this paper. The correlation matrixes of the six subgroups can be found in appendix D.

Further Implications from the Multiple Regression Analysis

The results of the multiple regression analysis support the implication from the correlation analysis. No comparability between the individual dimensions of readiness and resistance can be reached. It is evident that the antecedents, which account for the prediction of either one of the concepts, are very different to each other.

(31)

readiness and resistance so that these issues will be further analyzed to see whether these effects are comparable for the three dimensions of the concepts.

To test for moderation effects, as outlined by Baron and Kenny (1986), the standardized values for participation, OC, POS and self-efficacy have been computed, as well as the products of the possible combinations of the standardized variables. Performing multiple regression analysis with all relevant combination yielded no significant contribution of the product variables on the dependent variables, which were the three dimensions of readiness and resistance. The regression results of all combinations can be found in the appendix F. From this analysis it can be concluded that no moderating effects between the independent variables are evident within the sample.

The unexpected results can only be found in the regression models on all three dimensions of readiness. With regard to resistance the regression shows expected results that align with the findings of the correlation analysis. The intentional dimension of readiness shows positive correlations with all four independent variables in the correlation matrix, excluding OC this statement is also true for the emotional and cognitive dimension. Nevertheless, the results of the multiple regression analysis show that on the intentional dimension only self-efficacy and on the emotional and cognitive dimension only POS yields significant contributions to the model. As all independent variables (except OC and Self-efficacy) correlate strongly with each other, meditating effects can be assumed and have to be tested for.

Baron and Kenny state:

(32)

Intentional Readiness. Starting with intentional readiness, participation is the first variable to

be tested for being mediated by another independent variable. When performing a simple regression participation yields a beta of .303, which is significant at the .01 level. Entering POS or OC into the regression does not result in significant contribution of either of them, so that it can be concluded that no mediation is evident. Different results can be seen when adding self-efficacy to the model. Self-efficacy has a beta of .264 that is significant at the .01 level and the beta for participation drops to .222 and is now only significant at the .05 level. These findings prove a partial mediating of participation through self-efficacy with regard to intentional readiness.

The following independent variable that has been tested for mediation was POS. The regression of only POS on intentional readiness yields a beta of .261, which is significant at the .01 level. Entering self-efficacy into the model, in a second step the beta and the significance of POS drops to values of .193 and .053 respectively. Self-efficacy shows a beta of .292, which is significant at the .01 level. The regression shows that POS is partially mediated by self-efficacy. Following the same steps with OC instead of self-efficacy no mediation effects can be observed. Testing the mediating effect of participation the regression shows that POS is fully mediated by participation as its beta decreases to .085 with a significance of .512 and participation yields a beta of .266, which is significant at the .05 level. These results imply that POS is mediated to some degree by self-efficacy and by participation. Testing a model, which includes all three independent variables, the results show that self-efficacy fully mediates POS and participation. To establish a sound causal relationship the mediation effects of participation with regard to the relationship between POS and self-efficacy need to be investigated. The results show that participation fully mediates the influence of POS on self efficacy.

The final independent variable, which needs to be investigated, is OC. Testing for mediation effects of self-efficacy and POS do not yield any results. The regression of OC on intentional readiness results in a beta of .196, which is significant at the .1 level. Entering participation into the regression proves full mediation as the beta for OC becomes .120 and the significance is .245, whereas for participation the values are .276 and .009 respectively.

(33)

Figure 1

Antecedents of Intentional Readiness

Emotional Readiness. On the emotional dimension of readiness the correlation matrix does

not show any significant influence of OC. For this reason the following analysis of mediation concentrates solely on self-efficacy, POS and participation.

Starting with participation the regression analysis on emotional readiness results in a beta of .193, which is significant at the 0.1 level. Adding self-efficacy into the regression model supports the idea that participation is fully mediated by self efficacy. The values for participation decrease to a beta of .116 and a significance of .265, whereas self-efficacy yields a beta of .251 with a significance of .017. The regression analysis on emotional readiness for participation and POS results as well in a full mediation of participation by POS. The beta value for participation decreases to -.068 and becomes non significant, whereas POS yields a beta of .381, which is significant at the .01 level. The results show that the little influence participation has on emotional resistance is fully mediated by self-efficacy and POS.

The regression analysis for self-efficacy on emotional readiness results in a beta of .288, which is significant at the .01 level. The addition of POS into the regression analysis reveals a mediation effect that is turned around from what was observed for intentional readiness. For emotional readiness self-efficacy is partially mediated by POS, because the beta value decreases as well as the significance level. Partial mediation is evident, because the beta for self-efficacy still remains significant at the 0.05 level. Starting the regression with POS does not show mediation effects, as the beta of POS and its significance remains almost unchanged when entering participation and self-efficacy into the model.

(34)

Figure 2

Antecedents of Emotional Readiness

Cognitive Readiness. The variables, which influence cognitive readiness, are self-efficacy,

participation and POS. The following analysis concentrates on mediating effects regarding the before mentioned variables. The results show that self-efficacy is fully mediated by participation. The regression of self-efficacy on cognitive readiness yields a beta of .228, which is significant at the .05 level. By adding participation into the model the values drop to .163 and .119 respectively, whereas participation has a beta of .286, which is significant at the .01 level. Further expansion of the model through POS shows an ongoing mediation of participation by POS. the beta for participation decreases to .131 and becomes non significant and POS yields a beta of .257, which is significant at the .05 level.

The analysis shows a causal relationship for the antecedents of cognitive readiness. The model is shown in Figure 3 and explains 17.7% of the variance.

Figure 3

Antecedents of Cognitive Readiness

(35)

Behavioral Resistance. For behavioral resistance the evolving picture is very clear. The only

independent variable that shows an effect is self-efficacy and the results of the multiple regression analysis support this finding. For these reasons the model, which evolves from the sample with regard to behavioral resistance includes only self efficacy and explains 13.4% of the variance. The regression shows that self efficacy yields a beta of - .365, which is significant at the .001 level. Figure 4 shows this model.

Figure 4

Antecedents of Behavioral Resistance

Emotional Resistance. On the emotional dimension of resistance correlation analysis reveals

that only two independent variables have a significant impact. These two variables are self-efficacy and POS. Performing multiple regression analysis while entering the variable into the model one by one, starting with POS shows that POS is fully mediated by self-efficacy. The regression of POS on emotional resistance results in a beta of - .234, which is significant at the .05 level. When self-efficacy is added in the second step the beta of POS decreases to a value of - .128 and becomes non significant while self-efficacy yields a beta of - .444 and is significant at the .001 level. The resulting model represents a variance of 24.1% and is shown in Figure 5.

Figure 5

Antecedents of Emotional Resistance

Cognitive Resistance. As well as on the emotional dimension, on the cognitive dimension of

(36)

POS. The model for the regression analysis, which includes both independent variables yielded significant beta values for both of them and explained 20.8% of the variance. The model that evolves from this analysis is shown in Figure 6.

Figure 6

Antecedents of Cognitive Resistance

When comparing the models resulting from this analysis on their respective dimensions for readiness and resistance, major differences can be seen. So that again, the findings do not support either of the earlier stated hypotheses. All regression results are given in appendix E.

DISCUSSION

The underlying question of this research is, whether readiness and resistance can be seen as two extremes on a continuum of attitudes towards change. The implication of this hypothesis would be that it is enough to concentrate on either one of the concepts when being confronted with a change project. However, results show that this is not the case. Readiness and resistance differ substantially.

The concepts are defined by Armenakis (1993) and Oreg (2006) as having three distinct dimensions, namely intentional/behavioral, emotional and cognitive. Correlation results have shown that on either of these three dimensions readiness and resistance do have a strong negative influence on each other, but they do not correlate enough to say that they are counterparts to one another. Hypothesis 1 was not supported. Readiness for change and resistance to change have individual complex underlying concepts, which define how people behave when confronted with change in an organization.

(37)

process (Armenakis et al., 1993; Piderit, 2000; George and Jones, 2001; Oreg, 2006; Bouckenooghe et a., 2009), where the behavioral level of the resistance concept aims at behavior the individual has already shown in response to the change (Oreg 2006). The results show that this dimension produced the lowest correlation, which supports the idea that the difference between intentional readiness and behavioral resistance is the greatest, because they focus on different points in time. Nevertheless, correlation shows a high coefficient and supports the idea that there is a relationship between what employees want to do regarding the changes and what they show as behavior. This finding supports the idea of Armenakis et Al. (1993) that readiness is the precursor of either resistance to, or support for a change effort.

Emotional readiness and emotional resistance produce a stronger correlation coefficient, which means that greater comparability exists. Even though emotional responses were expected to be similar, as Armenakis et al. (1993) and Oreg (2006) define the dimension in the same way, this was not the case. What follows from this analysis, is that people can have distinct emotions, which differ with regard to readiness and resistance. The positive emotions associated with the change do not hold the balance with the negative emotions associated with the change. These findings imply that people can feel that the change is good, needed and some how refreshing, but still be afraid of the future, so that while being fully ready to take on the change, they can still show resistance as they deal with uncertainties and “a bad feeling”. Vice versa is it possible that employees are showing a high degree of resistance on the emotional level, stemming from uncertainty and fear, but still have some degree of readiness to take on the change as they might, as well, feel that the change is good and needed.

(38)

readiness, because he thinks that the changes are necessary for the organization to function in future environmental circumstances and demands.

The results clearly point into the direction that readiness and resistance influence each other, but they are not the two extremes on the continuum of attitude towards change. Further analysis has been performed to see whether the influence of the antecedents is similar for the two concepts on the comparable dimensions. It was assumed that all four independent variables should influence the level of readiness and resistance similarly. Analysis showed that readiness and resistance are not comparable to each other and that they have a complete different compilation of relevant antecedents

Self-efficacy is named in most studies as one of the major antecedents of readiness and resistance. The correlation results support this view as self-efficacy correlates with readiness (Ashford, 1988; Rafferty and Simmons, 2006) and resistance (Ellen et el., 1991) on all three dimensions. These findings where expected from the theory review. The comparison of the coefficients, however, revealed that the influence is different alongside the comparable dimensions of readiness and resistance. The influence on resistance is greater. A person with a high self-efficacy is less affected by the influence of uncertainty about and the fear of the change. Surely, this increases the readiness of employees to engage in change because they are convinced that they can handle the situation. However, the effect of uncertainty and fear seems to be stronger related to resistance than to readiness, because self efficacy has a greater impact on resistance. This finding supports the notion that self-efficacy reduces switching costs, which can mediate the influence on resistance and therefore the influence might be stronger (Kim and Kankanhalli, 2009). Nevertheless, it was not possible to confirm the hypothesis that readiness and resistance are comparable, because analysis has shown that self-efficacy has significantly different correlations with the concepts.

(39)

The research on the influence of POS on readiness and resistance was not supportive for all hypotheses. The relationship between POS and readiness was supported by the analysis and the assumed relationship with resistance was supported to some degree. For readiness these findings support the research of Eby et al. (2000) as well as of Rafferty and Simmons (2006). Employees, who perceive the organization as being supportive, are more ready for changes on all three dimensions. Perceived organizational support increases the readiness of employees to put effort in the upcoming changes and more behavior in support for the change is anticipated. It also increases the emotional and cognitive readiness. POS shows significant influences on resistance only on the emotional and cognitive dimension, so that with or without perceived support employees will show resisting behavior during the change process. The analysis has shown that the influence on the emotional and cognitive dimension differ significantly, so that again one can conclude that readiness and resistance have to be studied individually.

(40)

significant influence on resistance is further support on the notion that readiness and resistance are different concepts that interact, but are not counterparts.

Employees with a strong organizational commitment should be more willing to support the strategic decision of their company and they show less resistance to change and more readiness for change (Darwish, 2000; Iverson, 1996; Lau and Woodman, 1995). The results of the correlation analysis show that the assumed relationship cannot be proven from the research. Organizational commitment shows only a weak nearly non-significant correlation with the intentional dimension of readiness. The behavioral dimension of resistance is, however, not influenced by organizational commitment. Whether people protest or raise their concerns openly is not affected by their level of committed. The concept of commitment as defined by Meyer and Allen (1991) raises the idea that commitment as well as readiness and resistance can be split into three dimensions. These dimensions are defined as affective commitment, normative commitment and continuance commitment. Testing the correlations of these three distinct dimensions with readiness and resistance might reveal different results. Tests for moderation effects of organizational commitment on one of the before mentioned relationships did not yield any positive results either. From this analysis it can be said that the effects organizational commitment has on the attitudes towards changes are minimal and the concept should not be included into a prioritized analyses of action to be taken in change management.

In a further step of the analysis I was able to show that the regression models for readiness and resistance on the comparable dimensions, which consisted of the four tested independent variable, are significantly different from each other. I used the different models to compute a predicted outcome for readiness on the respective dimensions and correlated the resulting three variables. For all three dimensions the results of Steiger’s z (1980) to test for significant different correlation between interrelated variables showed that the models are different, so that it can be concluded that the concepts readiness and resistance are different from one another.

(41)

The model, which conceptualizes the antecedents of intentional readiness, includes all of the independent variables. The main contributor to readiness on this dimension is self efficacy. The other three variables are mediated. Comparing the conceptual model for readiness with the one for behavioral resistance the differences are very obvious. Self-efficacy is the only antecedent. The analysis of the conceptual models supports the idea that readiness and resistance are very different concepts with regard to their intentional/behavioral dimension. Nevertheless, self-efficacy seems to be the major contributor to behavior in change processes, whether this is intended or already shown. With a high self-efficacy employees might see the change as a challenge and are motivated by the new tasks and interesting upcoming dynamics.

Intentional readiness is influenced by a very specific causal relationship. POS and OC increase the perceived participation within the change process. Participation, in fact, has than a positive impact on the self-efficacy of the individual, which in the end increases their motivation and enthusiasm to engage actively in the change process. On the resistance measure, the mediating effects of self-efficacy do not take place, as Participation, POS and OC do not influence the construct. Only an increase in self-efficacy decreases resisting behavior as people are not so afraid of the new and future situation.

The conceptual model for emotional readiness includes participation, POS and self-efficacy. On the emotional dimension of readiness POS and self-efficacy contribute directly to readiness. Both of them mediate the influence of participation as, with regard to emotional readiness, higher perceived participation increase the self-efficacy and the perceived organizational support. A final implication from the analysis is that self-efficacy is partly mediated by POS. This finding is rather surprising as one could expect that greater support would result in a higher self efficacy, which then increases readiness, as it is with the intentional dimension. On the emotional dimension the implication is that people with a high self-efficacy might perceive the support by the organization as being stronger, because they do not have so much fear of the upcoming situation.

(42)

The conceptual model, which describes the antecedents of cognitive readiness, shows a straight and cascading causal relationship of the variables. POS is the main contributor with a positive influence on readiness. POS by itself mediates the effect of participation, which mediates the effect of self-efficacy. On the cognitive dimension it is evident that people with a strong self-efficacy feel more integrated into the process as they perceive higher participation. Through greater perceived participation they have the feeling that the organization supports them, which leads to a better cognitive sense making for the change and eventually to higher readiness.

Cognitive resistance is only influenced by POS and self-efficacy. No mediating effects are evident, so that both influence they way in which employees think about the change. More support and more self-efficacy lead to a more positive image of what the change will bring. Finally, the last two models are not comparable as well. Again, this supports the idea that the hypotheses of the research need to be rejected and one has to conclude that readiness and resistance are concepts, which are different and need to be dealt with individually.

CONCLUSION

The concepts of readiness and resistance seem to be very similar. Nevertheless they are not the same. Of course, strong relationships exist between the two concepts, but they differ in their implications for change efforts and they do not stem from the same contexts, nor are they influenced by change variables in the same way. It is therefore very important to look at both, readiness and resistance, when dealing with organizational change. The statement, that readiness precedents resistance, has truth to it. This implies that both concepts aim at different points in time of the change process. Readiness is an issue when change is anticipated and efforts should be made and actions should be taken to make employees ready for the change. When the process starts resistance is the concept that should be in the focus.