THE EFFECTIVNESS OF TRAINING AS A CHANGE
MANAGEMENT TOOL
A study at KPN
Master Thesis
MSc Business Administration, Specialization Change Management University of Groningen, Faculty of Economics and Business
3
THE EFFECTIVNESS OF TRAINING AS A CHANGE
MANAGEMENT TOOL
A study at KPN
Abstract
This research intends to investigate the effectiveness of training. The effectiveness of training has been measured by using behavioural outcome criteria. The research has been done at KPN a Dutch telecommunication company. The training that has been investigated had the purpose to change the subjective norm and attitude towards hospitality behaviour. This should have in the end the result that the employees behave in a more hospitable manner. In order to execute the measurement of the behavioural outcome, a scale had to be constructed in order to measure hospitality behaviour. The link between training and behavioural outcome could not be made in this research, however the factor principles of behaviour seems to increase due to training. Next to that a marginally significant increase in attitude due to training has been found. Since there is little research about the link between training and outcome criteria this research intends to open the door for further research that can uncover the relation between training and outcome criteria.
Keywords: Change management, training, attitude, hospitality, behaviour, Research Theme: Change Management
TABLE OF CONTENTS
INTRODUCTION ... 6
Organization / Context of research ... 7
Change at KPN ... 8
THEORY ... 9
Training ... 9
Theory of planned behaviour ... 11
5
Theoretical Implications, Limitations & Further Research ... 30
Operational implications ... 32
LITERATURE ... 34
APPENDIX A ... 37
6
INTRODUCTION
In today’s open market competition between organizations is about more than price and diversification. It is hard for a company to distinguish itself from competitors, since customers have almost unlimited choice and can easily compare different sorts of one product/service because information is free accessible through Internet. Another way to be different for a company is by providing a higher or more hospitable service standard than competitors in the same industry. According to Jorritsma & Wilderom (2012) an organization that improves its service level in order to gain competitive advantage often entails a cultural change strategy.
Employees, whom are part of such a change, change most of the time not by themselves. It is even harder for non-managerial employees to change to a new working environment. (Jorritsma, et al. 2012). Within change management many recipes for change have been described. Author’s like Kotter & Schlessinger. 2008; Burnes. 2004; Kotter. 1995 and Beer & Nohria (2000) describe how successful change should be accomplished. Most of these recipes are based on case studies, which describe a success story or failure. Other change literature tends to focus on factors that are important like resistance to change and communication (Ford, Ford & D’Amilio 2008; Szbla, 2007; Gilley, Dixon & Gilley, 2008.) Despite of this research 70% of change initiatives fail according to Beer & Nohria (2000). This could be due, to a lack of knowledge and research on very specific instruments of change management. As Jorritsma, et al. (2012) argue that the role of internal change agents and the communication and training they use are understudied.
7 Organization / Context of research
In order to measure the effectiveness of training within a cultural change program, data will be collected at KPN. Royal KPN N.V is a Dutch telecommunication company. Since the privatisation of the company, which was initiated in 1994 the organization struggles with its image. On the one hand the company is known as a stable, expensive but reliable service provider, on the other hand the organization has a bureaucratic and somehow customer unfriendly image among general public. This is a problem where a lot of large (telecom) companies struggle with nowadays. KPN wants to change this image and become the most customer-friendly organization within the Netherlands. To initiate this, the organization developed a strategy and three core values, which should be known, acted like and propagated by all of its employees. These values are: Personal, Simplicity and Thrust. Each of these values consists out of several subjects. Royal KPN N.V started as a state publicly owned company, which was property of the Dutch-state. It was formerly called Koninklijke PTT Nederland. Besides telecommunication services, this former company also controlled the national Dutch postal services. In 1994 the Dutch government began privatising the company. The telecommunication and postal service division were separated in two individual companies. In 2006 the Dutch government reduced its investment in KPN to 0.00%.
8 Change at KPN
In order to become the most customer friendly service organization within the Netherlands, KPN changed its strategy and with that the core values of the company. The change of focus in this research is within the retail division of KPN. In order to become the most customer friendly organization KPN wanted to realize a retail store, where everything is possible. These stores are called XL Stores. Within the XL store, customer friendliness and hospitality are the keywords. The stores should create a complete different ambiance and customer experience. In order to facilitate this, KPN invested in new working procedures, structure changes and changed the job description and the required skillset of employees. KPN has chosen for a strategy in which stores that were formerly so called business centres (aimed at businesses) will be turned into an XL store with retention of the current labour force. In order to become more customer friendly, the organization wants to create a hospitable culture / climate and wants the labour force to behave more hospital. A good example of this is the introduction of a hostess, which welcomes and guides customers who enter a KPN XL store. In order to become more hospital, the employees have to take part in three hospitality-training sessions.
These sessions have the purpose of changing the culture within the company to a more hospitable one. The result of this should be that the employees behave in a more hospitable / service manner (behaviour). The training focuses on sense making in order to change the attitude of employees towards hospitality. The training should help employees to look at a totally different way to service providing (cognitive). In summary the employees have to get a totally different feeling with their work, which should result in, different (more hospitable) behaviour. In order to accomplish this change, the organization strives to change the attitude and cognition of the employees by training them, in order to change their actual behaviour.
9
Is training an effective change management tool for KPN in order to change
employee behaviour?
In order to answer this question in the following chapter we will describe existing literature to make a proper proposition. In the next chapter we will describe the method we used to carry out this research. In the following chapter we present the results of our research. In the final chapter a discussion will follow and we will make recommendations for further research.
THEORY
Every change process is unique; there is a tremendous amount of approaches and types of change. Various researchers (Dunphy & Stace, 1993; Burnes, 2009; etc.) tried to define different types of change and place them in a framework or even made a recipe for change. This literature does however not give a clear picture how such a change should be initiated on the operational level and which tools are effective and how they should be used. Within this research we focus on the outcome criteria of the change, which will be measured according to employee behaviour and attitude towards hospitality behaviour. This research intends to explore if training employees is an effective tool in order to change actual behaviour.
Training
10
training is indeed an effective tool in order to change the attitude and behaviour. It is however hard to state that training is effective in general, since there can be used different types of measurement to measure training effectiveness and there are multiple types of training methods. According to Arthur, Edens, Bell & Bennet (2003) based on (Gagne, Briggs, & Wagner, 1992; Rasmussen, 1986; Schneider & Shiffrin, 1977) types of training can be brought back to three typologies. These typologies are cognitive, interpersonal and psychomotor training. Cognitive and task related training are related to how somebody thinks, idea generation, understanding, the knowledge requirements for a job and problem solving. Interpersonal skills are about interacting in a group, or interacting with customers, they encompass a wide variety of skills including leadership skills, communication skills, conflict management skills, and team-building activities. Psychomotor training encompasses physical or manual activities.
Arthur, Edens, Bell & Bennet (2003) point out that it depends on the match between training method and what needs to be trained if a training method is effective. This is in line with Wexley and Latham (2002) who state that training effectivity depends on the subject, of what needs to be learned and the used methods. Wexley and Latham (2002) and Arthur, Edens, Bell & Bennet (2003) point out that there has been however very little, or no research that directly assesses the effect between a specific training method and the effectiveness of training. They assume however that the combination of cognitive and interpersonal training had the best effect on training effectiveness. At KPN a combination of behavioural and interpersonal skill training have been used.
11
do actually show the effectiveness of training, because they wonder how trainees feel about training can actually tell anything about the effect of training. Arthur, Edens, Bell & Bennet (2003) state furthermore that behavioural and learning criteria are conceptually linked, however according to them researchers have failed to empirically demonstrate the relationship between training and these criteria (Colquitt, LePine & Noe. 2000). This is because the environment can influence the transfer or use of trained skills. Arthur, Edens, Bell & Bennet (2003) found that it is relative how effective training is. Their results suggest that the combination of the used training method, skill or task characteristics that are trained and the choice of how the training is evaluated influence the observed effectiveness of a training program. They concluded furthermore by comparing learning criteria with subsequently result and behavioural criteria that the effect size for learning criteria were higher than for behavioural and result criteria. This finding assumes that it is difficult to find a connection between training and behaviour. This might be the reason that there is not much research, which investigates the relation between learning and behavioural outcomes. We think however that due to the lack of such research this research can help to investigate such a possible relation. In the next paragraph we will investigate and propose why training could be an effective tool to change behaviour. This will be done through investigating the underlying process of how behaviour can be changed.
Theory of planned behaviour
12
Figure one displays the theory of planned behaviour. Intention is the central
factor in the theory of planned behaviour. According to Azjen (1991) “Intentions are
assumed to capture the motivational factors that influence behaviour.” Azjen (1991)
furthermore states that the higher an intention is, the higher the probability, that someone will exert the behaviour. Azjen (1991) states that intention can only find expression in actual behaviour when there is volitional control. Azjen (1991) explains volitional control as: “if the person can decide at will to perform or not
perform the behaviour”. He however notices that exerting the actual behaviour is also
influenced by the actual control of behaviour (time, money, skills, cooperation, etc.). The theory of planned behaviour assumes that the intention to execute behaviour or not execute the behaviour is the leading principle in order to decide whether the behaviour will be executed or not.
According to the theory of planned behaviour there are three determinants, which influence the intention (see figure.1). These three determents are: perceived behavioural control, attitude and subjective norm. These three determinants can, according to the theory of planned behaviour predict intention. Since the training at KPN focuses on attitude and cognition in order to change behaviour, we will focus in our research on attitude and the subjective norm, because these are the two determinants the training focuses on.
13
the actual behaviour that is influenced by the environment. At KPN the training is focused at changing the subjective norm through cognition. By explaining why
employees should behave different and what the results might be. The attitude
toward the behaviour is the degree to which a person has favourable or unfavourable evaluation or appraiser of the actual behaviour (Azjen, 1991). In other words the attitude is about an individual’s positive or negative evaluation of the behaviour in question. The attitude is another subject of focus during the training. The training at KPN intends to change the attitude towards hospitality behaviour.
Armitage & Conner (2001) found that 39% of the variance in intention can be explained by attitude, subjective norm and perceived behavioural control. They state that the importance of the variables; attitude, subjective norm and perceived behavioural control are relative. They state that the power of these variables can vary across situations. Within this research we focus on behaviour and attitude. Since we have chosen to measure on the output criteria we measure the change in behaviour instead of the change in intention. According to the behavioural criteria we use, the training would be effective when the behaviour changes and not when the intention of the respondents changes and the behaviour not. For that reason the first hypothesis is: • H1: Employees who participated in a training session will show more
hospitable behaviour than employees who did not participate in training. As attitude towards the (new) behaviour is one of determinants of actually performing that behaviour, and since the training given within KPN is focused on changing behaviour and attitudes, we also hypothesize the following:
• H2: Employees who participated in a training session will have a more positive attitude towards hospitality than employees who did not participate in training.
14
planned behaviour since the subjective norm refers to the perceived social pressure to perform the actual behaviour (Azjen,1991). As mentioned before the subjective norm is someone’s perception about the actual behaviour that is influenced by the environment. The expectation that one is given a hospitality training, could change the perceived social pressure to perform in a hospitable behaviour and therefore the perception of the desired behaviour. For that reason, we propose:
• H3 Employees who knew that they were to be participating in the training sessions will show more hospitable behaviour than employees who did not participate in training and thus, did not expect to be in a training therefore had no expectations.
So in fact: we assume that expectations to be in a training play a role in the actual effectivity of the training, so we will use the prior measurement of hospitality behaviour as a control variable for the actual effect (post-training measurement) of training on this behaviour.
Now we proposed our hypothesis, in the next chapter we will operationalize hospitality behaviour, since this is the dependent variable within this research.
Hospitality
15
discovering the culture of restaurants and hotels. An example of this is research by Woods (1989). Woods laid the foundations of hospitality research, by examining five restaurants. He states: “that hospitality culture is unique in the fact that the product
and the service are integrated to provide an experience for the customer”. Kemp and
16
precision in their work. These attributes include, detail oriented and precision. The last factor Composure is related with someone’s general behaviour. The attributes of Composure include, composure and having low anxiety.
Since Azjen (1991) argues that attributes are linked to behaviour we used these attribute item list of the hospitality culture scale as the basis for measuring behaviour. In the next chapter the method used to perform this research will be described.
METHOD Research Design
This research intends to evaluate the effectiveness of training on behaviour and attitude. Behaviour has been measured with the use of a self-composed survey, which is based on the hospitality culture scale of Dawson, et al. (2011). In order to locate differences prior and post to training the behaviour of the respondents needed to be measured twice. The first measurement was a month before the actual training and the second measurement was a month after the actual training. These measurements where performed at three locations that were part of the change and participated in the training. During both measurement moments the respondents had to self evaluate their behaviour with the use of a survey.
17 future.
The second measurement tool was attitude towards hospitality behaviour, since attitude is according to literature a predictor of actual behaviour as has been explained in the theoretical framework (Azjen, 1991). The attitude was only measured at the second measurement moment of the target and control group. Attitude has been measured with the use of a self-composed survey, which is based on the hospitality culture scale of Dawson, et al. (2011).
Due to the limited time frame to do two complementary measurements, a low sample size and the possibility that respondents over estimate themselves during self-reporting there has been chosen to use qualitative research as well. Three managers of the departments that were included in this research were asked to evaluate the effectiveness of training on their team. The interviews have been done with three managers of the departments that were involved. They were asked to give their team a score per item, after and before the training session. The same survey that has been used for the self-reporting surveys, during measurement moment two has been used during the interviews.
Sample
The respondents were chosen from three locations that were involved in the change process. The total workforce of these locations consisted out of 39 persons of which 71.8% were man and 28,2% were woman. The average age of this group was 33.33. All employees at these locations were approached. 16 persons filled out the survey completely, at measurement one and 15 persons filled out the survey at measurement two. The average age at measurement one was 31.1 and at measurement two 28.9. At measurement one 73.7 % of the respondents were man and 26.3% were woman. At measurement two 68,4% of the respondents were man and 31.6% were woman.
18
out the survey completely. At measurement one, 71.4% were man and 28.6% were woman. At measurement two, 83.3% were man and 16.7% were woman. The average age at measurement one, was 29.3 and at measurement two, 32.1.
The qualitative interviews can be labelled as structured interviews (Blumberg, et al. 2008). The very detailed interview guide that has been used was the same as the item list that has been used in order to measure the behaviour and attitude with the use of self-report through a seven-point likert scale. Demographic data of the interview respondents are not provided, in order to guarantee their anonymity. Concluding can be said that these numbers are representative for the population. When we compare these background variables with demographic data of the organization it is obvious that the representation of our sample fits the company profile whereas 24% of the workforce consists out of woman.
Data collection
At first, data was collected with the use of an online-survey, but because of a low response rate the survey has also been carried out on paper. The data has been collected anonymous in order to give respondents the possibility to give real answers instead of socially desirable answers. Due the anonymity of the surveys, it was not possible to compare the same respondents during measurement one and two. For that reason only the means between the training and the control group of measurement one and two could be compared.
19
The surveys have been developed in Dutch, since the workforce of KPN is native Dutch speaking. After developing the scale, the item list was reviewed with the use of employees that were no part of the sample or control group. By reviewing the survey, difficult questions or questions that could be wrong interpreted could be deleted or reformulated.
By using a survey the employees were asked to evaluate themselves. The used scale to measure attitude and behaviour was a seven-point likert scale. The respondents could indicate how much they agreed upon an item by using a seven point likert scale. The scale had a range from totally disagree to totally agree. There has been chosen for a seven point likert scale instead of a five-point likert scale in order to exclude the possibility to give a neutral answer. Next to that the use of seven answer possibilities gives a broader spectrum to measure small changes during the different moments of measurement than a smaller scale.
Scale construction
20
Since there was no scale available that measured hospitality behaviour, we argue that attributes that indicate a hospital culture are logically related to hospitality behaviour (Azjen. 1991). Within their research (Dawson, et al. 2011) found with the use of factor analysis six factors of personal attributes within hospitality culture: principles, propitiousness, leadership, risk taker, accuracy and composure. We included all items of these factors except the factor leadership, since our respondents were sales employees and are not excepted to exert leadership behaviour as described. The item list can be found in Appendix A. The attitude item list was also composed wit the use of arbitrary scaling. According to Azjen (1991) attitude toward the behaviour is the degree to which a person has favourable or unfavourable evaluation or appraiser of the actual behaviour (Azjen, 1991). We measured this by measuring the favourable or unfavourable evaluation of hospitality attributes from hospitality literature (Dawson, et al. 2011) These attributes included helping guests celebrate milestones in their lives, creating memories for guests, creating a home away from home for guests, developing relationships with customers, having a high percentage
of repeat guests and considering “to be ‘of service’ a noble profession. The used item
list can be found in Appendix A.
Factor analysis
21
components we believed to exist are collared green and white in the table, which can be found in Appendix B. The squares that are collared red show that the items score higher on other components than we assumed. The outcome of both factor analyses could assume that there are smaller factors than the four we assumed to exist. We have however chosen to use the four factors we assumed to exist and found by Dawson, et al (2011). We use these factors for further analysis since we translated personal attributes and behaviours and preserved comparable items in our survey. The problem with the factor analysis we performed is, that the sample within this research is to small to perform a reliable factor analysis. This might explain the differences between the two factor analyses. In order to measure the reliability of the scale, the cronbach’s alpha has been used. According to Hair, Andersone, Tatham and Black (1998) a cronbach’s alpha between the values of 0.60 to 0.70 is the lower limit of acceptability. The crobach’s alpha of the behavioural item list at measurement one was 0.87 and at measurement two 0.92, this is sufficient since the score is >0.70. We deleted one item; item 10, from the behavioural item list, since this item reduced the
cronbach’s alpha >0,40 at measurement two.
There was also a factor analysis performed for the attitude item list. Attitude was only measured at the second measurement moment, because of that we had to perform one factor analysis. The outcome of this factor analysis can be found in Appendix B. The outcome shows that the attitude item lists consist out of one component. So there can be concluded that the attitude item list is consistent. The cronbach’s Alpha for the attitude item list is 0.856, this is sufficient since the score is >0.70.
Data analysis
22
unequal variances outcome of the independent sample T test. When the P>0.05 (the significance level) the output equal variances assumed should be used. When P<0.05 the output equal variances not assumed should be used. Since our hypothesis is a one tailed test the P value had to be divided by two to gain our test results. In order to test the hypotheses, new variables were computed. The variables were computed by calculating the mean of the different items that include behaviour and attitude. Next to that we tested with the factors, the factors were calculated using the same procedure.
The interviews were analysed by calculating the mean scores for the variables and by comparing the difference between the results of the three interview respondents. The qualitative data was compared and comparative statements of the three respondents were filtered out and included in the results.
RESULTS
In this chapter the results of this research will be described. In the first paragraph the descriptive statistics and the results of our hypothesis testing will be presented. In the second paragraph the results of the interviews will be presented.
Hypotheses Testing
In order to test the hypotheses an independent sample T test has been used. The independent sample T test has been used to test the following hypotheses:
• H1: Employees who participated in a training session will show more hospitable behaviour than employees who did not participate in training. • H2: Employees who participated in a training session will have a more
positive attitude towards hospitality than employees who did not participate in training.
23
Hypothesis three
In order to test hypotheses three (Employees who knew that they were to be
participating in the training sessions will show more hospitable behaviour than employees who did not participate in training and thus, did not expect to be in a training and therefore had no expectations) the means of the behaviour prior to the
training session of the target group and the means of the first moment of measurement of behaviour of the control group have been compared. The results can be found in table 1. The means of the behaviour as well as the behavioural factors have been compared.
Table 1: descriptive statistics for hypothesis 3
Wel of geen training / verwacht
training N Mean Std. Deviation
Std. Error Mean Behaviour NT* 14 5,6667 ,56226 ,15027 T** 16 5,8287 ,40047 ,10012 Principles NT* 14 6,1825 ,58569 ,15653 T** 16 6,2500 ,58584 ,14646 Presenation NT* 14 5,6883 ,59208 ,15824 T** 16 5,9034 ,60436 ,15109 Risk NT* 14 3,9524 1,59517 ,42633 T** 16 4,5208 1,00347 ,25087 Accuracy NT* 14 5,4286 1,05351 ,28156 T** 16 5,4688 ,82601 ,20650 Composure NT* 14 6,0357 ,81958 ,21904 T** 16 5,8438 ,67623 ,16906
*No Training, **Participated in Training
24
Table 2: Independent Samples Test H3
Levene's Test for Equality of Variances
t-test for Equality of Means
F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Behaviour Equal variances assumed 3,378 0,077 0,918 - 28 0,367 -0,16204 0,17653 0,52365 - 0,19958 Equal variances not assumed 0,897 - 23,15 0,379 -0,16204 0,18057 0,53544 - 0,21136 Principles Equal variances assumed 0,009 0,925 0,315 - 28 0,755 -0,06746 0,21437 0,50658 - 0,37166 Equal variances not assumed 0,315 - 27,475 0,755 -0,06746 0,21437 0,50695 - 0,37203 Presenation Equal variances assumed 0,182 0,673 0,982 - 28 0,335 -0,2151 0,2191 -0,6639 0,23371 Equal variances not assumed 0,983 - 27,616 0,334 -0,2151 0,21879 0,66355 - 0,23335 Risk Equal variances assumed 1,981 0,17 1,184 - 28 0,246 -0,56845 0,48007 1,55184 - 0,41493 Equal variances not assumed 1,149 - 21,344 0,263 -0,56845 0,49466 1,59615 - 0,45924 Accuracy Equal variances assumed 1,287 0,266 0,117 - 28 0,908 -0,04018 0,34346 0,74373 - 0,66337 Equal variances not assumed 0,115 - 24,582 0,909 -0,04018 0,34917 0,75993 - 0,67957 Composure Equal variances assumed 0,519 0,477 0,703 28 0,488 0,19196 0,27309 0,36743 - 0,75136 Equal variances not assumed 0,694 25,316 0,494 0,19196 0,27669 0,37754 - 0,76147 Hypothesis one
In order to test hypothesis one (employees who participated in a training
session will show more hospitable behaviour than employees who did not participate in training) the means of the behaviour post to training of the target group and the
25 risk taker.
Table 3: Group Statistics Hypothesis 1
Training / No Training N Mean Std. Deviation Std. Error Mean
BehaviourT2 No Training 12 5,8704 ,49224 ,14210 Training 15 5,9136 ,65371 ,16879 PrinciplesT2 No Training 12 5,8561 ,38852 ,11216 Training 15 6,2296 ,63893 ,16497 PresentationT2 No Training 12 5,7431 ,45982 ,13274 Training 15 5,8424 ,77098 ,19907 RiskT2 No Training 12 5,0417 1,55882 ,44999 Training 15 4,8000 1,28360 ,33142 AccurancyT2 No Training 12 6,0417 ,75252 ,21723 Training 15 6,1667 ,85912 ,22183 ComposureT2 No Training 12 6,0417 ,75252 ,21723 Training 15 6,3000 ,79732 ,20587
In order to test whether these higher mean scores are a significant difference, an independent sample T test has been carried out between the behaviour post to the training session of the target group and the second time of measurement of the control group. We tried to test this hypothesis with a Manova, but could not gain reliable results since our data set was to small to perform such a test. Whereas with a Manova the behaviour at measurement one could be used as covariance this is not possible with a t-test. However, we chose to use the variables of the measurement post to training session of the target group and of the second measurement in time of the control group to test our hypothesis. We chose not to correct the measures of the second measurement moments, because we believe that the rejection of hypothesis three shows that there are no significant other variables than training that influence
the behaviour. Hypothesis three showed that there is no significant difference between
the target and control group at measurement one.
26
by two to gain our test results. These results suggest that training does not have an
effect on the behaviour of employees.
Table 4: Independent Samples Test Hypothesis 1
Levene's Test for Equality of
Variances t-test for Equality of Means
F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper BehaviourT2 Equal variances
assumed ,461 ,503 -,190 25 ,851 -,04321 ,22779 ,51235 - ,42593 Equal variances
not assumed -,196 24,936 ,846 -,04321 ,22064 ,49768 - ,41126 PrinciplesT2 Equal variances
assumed 3,143 ,088 1,776 - 25 ,088 -,37357 ,21037 ,80683 - ,05969 Equal variances
not assumed 1,873 - 23,534 ,074 -,37357 ,19949 ,78572 - ,03858 PresentationT2 Equal variances
assumed 3,263 ,083 -,393 25 ,698 -,09937 ,25276 ,61993 - ,42119 Equal variances
not assumed -,415 23,344 ,682 -,09937 ,23926 ,59392 - ,39518 RiskT2 Equal variances
assumed ,433 ,516 ,442 25 ,662 ,24167 ,54660 ,88408 - 1,36742 Equal variances
not assumed ,432 21,256 ,670 ,24167 ,55887 ,91971 - 1,40305 AccurancyT2 Equal variances
assumed ,199 ,660 -,397 25 ,695 -,12500 ,31524 ,77424 - ,52424 Equal variances
not assumed -,403 24,753 ,691 -,12500 ,31048 ,76477 - ,51477 ComposureT2 Equal variances
assumed ,225 ,639 -,857 25 ,399 -,25833 ,30129 ,87885 - ,36218 Equal variances
not assumed -,863 24,257 ,396 -,25833 ,29929 ,87568 - ,35901
27
Hypothesis two
In order to test hypotheses two (Employees who participated in a training
session will have a more positive attitude towards hospitality than employees who did not participate in training) the means of the attitude post to the training session of the
target group and the attitude at the second measurement of the control group have been compared. The descriptive statistics can be found in table 5.
Table 5: descriptive statistics for hypotheses 2
Training or not N Mean
Std. Deviation Std. Error Mean Attitude No Training 12 5,9333 ,91585 ,26438 Training 15 6,3733 ,75163 ,19407
28 Interviews
Since the survey was developed to measure the outcome variables with the help of self-report. We have also interviewed the managers of the three departments, which took part in the training program. By interviewing the managers we tried to overcome self-reporting bias and assess the training program from another perspective. Next to that, interviews gave the opportunity to ask if the potential measured effect was caused because of the training or due to other variables. The results can be found in table 7 on the next page.
Table seven shows that there is difference among the three locations according to the direct managers of either location when it comes to behaviour. The results show that according to all respondents, hospitality behaviour increases a bit. The extend to which the mean score increases at the different locations differs however. The difference in mean score between the different locations when it comes to the attitude towards hospitality shows an increase at all locations. The extend of the increase differs however between the locations.
Table 6: Independent Samples Test H2
Levene's Test for Equality of
Variances t-test for Equality of Means
29
Table 7: Qualitative results
Location: 1 2 3 Prior to training Post to training Difference Prior to training Post to training Difference Prior to training Post to training Difference Principles 4,56 4,56 - 6,00 6,67 0,67 4,33 5,67 1,33 Propitiousness 4,91 5,09 0,18 5,73 6,18 0,45 5,00 6,27 1,27 Risktaker 4,33 4,67 0,33 4,00 4,00 - 7,00 6,67 0,33- Accuracy 5,00 4,50 0,50- 4,00 4,00 - 6,00 6,00 - Composure 5,00 5,00 - 5,50 7,00 1,50 6,50 6,00 0,50- Behaviour 4,74 4,81 0,02 5,48 6,15 1,12 5,19 6,07 2,27 Attitude 5,40 5,80 0,40 5,50 7,00 1,50 2,80 6,20 3,40 DISCUSSION
The goal of this research was to investigate whether training is an effective tool in order to change employee behaviour. The research question this research tries to answer was for that reason: Is training an effective change management tool for
KPN in order to change employee behaviour? This chapter tries to give an answer to
this research question.
30
behaviour. There are more variables that influence the actual behaviour than attitude. Ajzen (1991) model does for example not include variables like the physic possibility to exert certain behaviour. Next to that there is a gap between intention and behaviour. Azjen & Fishbien 2005 explain this as literal inconsistency. This means that people act different than they say they will do. This can simply be because people forget to act up on the new learned behaviour or because the external environment limits them to act upon the new learned behaviour, because (for example) the employees liked the post-training environment more (Azjen & Fishbien 2005).
The results of the qualitative research show a small increase in attitude and behaviour, it is however not clear if these changes are due to the training or other factors. The difference between the different locations could be explained by the different managers themselves, since Jorritsma & Widerom (2012) concluded that the role and support of internal change agents, including the communication and training they offer are understudied, while they are important elements in any organizational culture change. The managers themselves have part in the change program as internal change agents; they have direct influence on the employees, since they directly assess and steer the employees. From the results of the quantitative study we can also not conclude that training is an effective change management tool for KPN in order to change employee behaviour.
Theoretical Implications, Limitations & Further Research
31
there were changes in intention. Researching this relation would deliver more information about the effect of training and why the behaviour did not change.
This research has been conducted within one organization; KPN. In order to execute a reliable research, we used as well for the target and control group, respondents from multiple working locations. Despite the fact that within our research, we only collected data within KPN, does not make our result solely applicable to KPN. As mentioned in the introduction: providing a higher or more hospitable service standard than competitors in the same industry is a known strategy. Many companies throughout many industries use such a strategy, in that sense the situation at KPN is not a very specific one. For that reason the results could be generalizable and the results could be replicated throughout other companies. However, the sample we used to perform this research was very small, in order to test our results in further research it would be an addition to use a larger sample size in order to increase the reliability and test the results we found.
32
moments in time at the control group. The attitude toward behaviour was only measured post to training and at the second moment in time at the control group. For further research it would be better to measure attitude prior and post to training and at the first and second measurement of the control group. With such a method the first measurements could be used as a control variable in order to test if the change in attitude is a direct effect of the training. We chose to use a self-composed attitude survey list, based on the hospitality culture scale. We did this since we added the variable attitude at the second measurement of this research. There are however existing measurement instruments to measure attitude. For further research that investigates the same relations as in this research we would recommend to use an existing and proven method to measure attitude.
Within this research we made use of a control group, the respondents of the control group worked across other locations than the respondents of the target group. For further research it would be better to use respondents that do not follow a training program but work on the same location and conditions as the respondents of the target group. This would be a more reliable measurement, since external influences would be excluded in a better way. Another limitation of this study is that we analysed our data with the use of mean scores per variable. For further research it would be wise to compare the same respondents of measurement one and two. By doing so there is a possibility to analyse, if there are changes per respondent and is it possible to monitor if the sample prior and post to training are the same.
Operational implications
33
however not mean that there is not an effect. In order to make training programs in future more effective, some knowledge from existing literature could however be used. Before trying to change attitudes it should be clear which attitude and behavior should be changed and into what these should be changed. At KPN it was not clear what exactly hospitality was and what the current behavior and attitude towards the behavior was (Azjen. 2006). In the future it would be wise to have a clear picture of the current and desired situation.
34
LITERATURE
Ajzen, I. (1988). Attitudes, personality and behaviour. Chicago: The Dorsey Press. Ajzen, I. (1991). Theory of planned behavior. Organizational behavior and human
decision processes, 50(2), 179-211.
Ajzen, I. (2006). Behavioral Interventions Based on Theory of Planned Behavior. Ajzen, I., & Fishbein, M. (2005). The influence of attitudes on behavior. In
D. Albarracín, B. T. Johnson, & M. P. Zanna (Eds.), The handbook of
attitudes (pp. 173-221). Mahwah, NJ: Erlbaum.
Arthur, W., Bennett, W. Edens, P.S., & Bell, S.T. & (2003). Effectiveness of training in organizations: A meta-analysis of design and evaluation features. Journal of Applied Psychology, 88, 234-245.
Armitage, C.J., & Conner, M. (2001) Efficacy of the theory of planned behaviour:
A meta analytic review British Journal of Social Psychology, 40 (4), pp.
471-499.
Beer, M. & Nohria, N. 2000. Cracking the code of change. Harvard Business Review,
78:133-141.
Blumberg, B., Cooper, D. R., & Schindler, P. S. (2008). Business research methods.
2nd European edition. New York: McGraw-Hill.
Gagne, R. M., Briggs, L. J., & Wagner, W. W. (1992). Principles of instructional
design. New York: Harcourt Brace Jovanovich
Comrey AL. A First Course in Factor Analysis. New York: Academic Press, Inc; 1973.
Dawson, M., Abott, J., Shoemaker, S. (2011). The Hospitality Culture Scale: A measure organizational culture and personal attributes. International
35
Grossman, R., & Salas, E. (2011). The transfer of training: what really matters. International Journal Of Training & Development, 15(2), 103-120
Hair J, Anderson RE, Tatham R.L, Black WC. (1995) Multivariate data analysis.
4th ed. New Jersey: Prentice-Hall.
Hair, J. E., Anderson, R. E., Tatham, R. L. and Black, W. C. (1998). Multivariate
Data Analysis. 5th Edition. New Jersey: Prentice-Hall.
Jones, G. (2010). Organization Theory, Design and Change. 6th edition. New Jersey:
Pearson.
Jorritsma, Y., & Wilderom, C. (2012). Failed culture change aimed at more service provision: a test of three agentic factors. Journal of Organizational Change
Management, 25(3), 364-391.
Kemp, S., Dwyer, L., 2001. An examination of organizational culture: the Regent Hotel, Sydney. International Journal of Hospitality Management 20, 77–93. Kirkpatrick, D. L. (1959). Techniques for evaluating training programs. Journal of the
American Society of Training and Development, 13, 3–9.
Orbell, S., Hodgkins, S., & Sheeran, P. (1997). Implementation intentions and the theory of planned behavior. Per- sonality and Social Psychology Bulletin,
23, 945–954.
Rasmussen, J. (1986). Information processing and human–machine inter- action: An
approach to cognitive engineering. New York: Elsevier.
Salas, E., & Cannon-Bowers, J.A. (2001). The science of training: A decade of
progress. Annual Review of Psychology, 52(1), 471–499.
Salas, E. and Stagl, K. C. (2009), ‘Design Training Systematically and Follow the Science of Training’, in E. Locke (ed.), Handbook of Principles of
36
Schneider, W., & Shiffrin, R. M. (1977). Controlled and automatic human information processing: I. Detection, search, and attention. Psychological
Review, 84, 1–66.
Slotboom, A. (2008). Statistiek in woorden, 4 edition .Groningen: Noorhoff uitgevers B.V.
Tiggerlaar, B. (2013, Juli 07). Waarom floppen veel veranderingen?.
NRC Handelsblad.
Tepeci, M., Bartlett, A., 2002. The hospitality industry culture profile: a measure of individual values, organizational culture, and person organization fit as predictors of job satisfaction and behavioural intentions. International Journal
of Hospitality Management 21 (2), 151–170.
Van Buren, M. E., & Erskine, W. (2002). The 2002 ASTD state of the industry report.
Alexandria, VA: American Society of Training and Development.
Wexley, K. N., & Latham, G. P. (2002). Developing and training human resources in
organizations (3rd ed.). Upper Saddle River, NJ: Prentice.
Williams, B., Brown, T., and Onsman, A. (2010) "Exploratory factor analysis: A five-step guide for novices," Australasian Journal of Paramedicine: Vol. 8: Iss. 3, Article 1.
37 APPENDIX A
Item list Hospitality Behaviour
1. Gedurende het dagelijks contact met klanten handel ik betrouwbaar. 2. Gedurende het dagelijks contact met klanten ben ik extreem eerlijk. 3. Gedurende het dagelijks contact met klanten handel ik altijd integer. 4. In het contact met klanten vind ik eerlijk handelen erg belangrijk.
5. Ik sta bij klanten bekend als iemand die zich aan gedane afspraken houdt. 6. Ik blijf gedurende mijn dagelijkse werkzaamheden altijd trouw aan mijn eigen
normen en waarden.
7. Klanten zouden mijn houding en gedrag als Loyaal omschrijven.
8. Ik probeer het goede voorbeeld te geven en klant verwachtingen te overtreffen.
9. Ik ben tevreden met mijn handelswijze richting klanten.
10. *Klanten zouden mijn houding en gedrag als betrouwbaar omschrijven.
11. Klanten en collega’s zouden mijn houding en gedrag beschrijven als vriendelijk en warm.
12. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik graag met mensen werk.
13. Klanten en collega’s zouden mijn houding en gedrag beschrijven als vrolijk. 14. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik open
ben.
15. Klanten en collega’s zouden mijn houding en gedrag beschrijven als zorgzaam en empatisch.
16. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik cultuur bewust ben.
17. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik graag anderen bedien.
18. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik meer sociaal dan onafhankelijk ben.
19. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik anderen vertrouw.
20. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik op de gemeenschap gefocust ben en niet op individueel niveau.
21. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik een positief persoon ben.
22. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik risico’s neem.
23. Klanten en collega’s zouden uit mijn houding en gedrag beschrijven als avontuurlijk.
24. Klanten en collega’s zouden uit mijn houding en gedrag afleiden dat ik gewogen beslissingen / risico’s neem.
38
26. Klanten en collega’s zouden uit mijn houding en gedrag tijdens een gesprek afleiden dat ik probeer zo precies en gedetailleerd mogelijk te zijn.
27. Klanten en collega’s zouden uit mijn houding en gedrag tijdens een gesprek afleiden dat ik kalm ben.
28. Klanten en collega’s zouden uit mijn houding en gedrag tijdens een gesprek afleiden dat ik besluitvaardig ben.
Attitude
1. Ik vind het fijn om klanten te helpen bij het maken van keuzes die invloed hebben op hun dagelijks leven.
2. Ik vind het fijn om een klant een positieve beleving te geven wanneer deze ons filiaal bezoekt.
3. Ik vind het fijn om een zakelijke relatie aan te gaan met klanten namens mijn filiaal.
4. Het filiaal waarin ik werk heeft een groot aantal trouwe / terugkerende klanten.
5. Ik vind service verlenen het mooiste werk dat er is.
39 APPENDIX B FACTOR ANALYSIS
Rotated Component Matrix Behaviour Measurement at Time 1 Component 1 2 3 4 5 6 7 1 ,505 ,046 -,090 ,105 ,298 ,001 ,604 2 ,692 -,244 -,154 -,362 -,014 ,293 ,234 3 ,818 -,102 ,139 ,260 ,071 -,023 -,180 4 ,820 -,082 ,024 ,008 -,010 ,070 ,122 5 ,807 ,020 ,155 ,144 ,236 -,080 -,053 6 ,745 ,228 -,107 ,233 ,161 -,229 ,171 7 ,398 -,185 ,105 ,783 ,040 ,161 -,156 8 ,701 ,031 ,117 ,018 -,058 ,364 ,095 9 ,728 ,258 ,147 ,291 -,106 -,131 -,030 11 ,052 ,720 ,063 -,051 -,198 ,247 ,370 12 ,098 ,801 ,107 ,004 -,070 -,139 ,003 13 -,217 ,774 ,352 ,073 ,224 ,025 -,108 14 ,046 ,748 -,014 ,152 ,173 ,371 ,098 15 -,006 ,324 -,151 ,422 ,166 ,690 ,124 16 -,016 ,141 ,480 -,073 -,069 ,693 -,220 17 ,514 ,285 ,365 ,164 ,227 ,451 -,054 18 ,140 ,145 ,087 ,869 -,076 -,009 ,078 19 ,393 ,435 ,397 ,052 -,306 ,037 -,056 20 ,324 ,191 -,086 ,500 ,166 ,319 -,531 21 ,103 ,588 ,333 ,420 ,229 ,138 -,110 22 ,282 -,051 ,816 -,065 ,148 ,143 ,129 23 -,029 ,193 ,843 ,142 -,013 -,077 ,166 24 ,066 ,367 ,727 ,129 -,108 ,104 -,258 25 ,197 ,055 ,076 -,032 ,786 ,084 ,357 26 ,054 ,048 -,029 ,063 ,830 -,009 ,104 27 ,049 ,075 ,081 -,053 ,349 -,011 ,747 28 -,072 ,577 ,020 -,192 ,465 ,284 -,221
40
Rotated Component Matrix Behaviour Measurement at Time 2
Component 1 2 3 4 5 6 7 8 1 ,303 ,551 -,012 -,152 -,054 -,406 ,352 ,193 2 ,392 ,324 ,367 ,215 -,314 ,367 ,380 -,105 3 ,813 ,290 ,054 -,050 ,182 ,008 -,020 ,145 4 ,360 ,825 -,121 ,064 -,161 ,112 ,106 ,108 5 ,743 ,379 -,006 ,135 ,128 ,009 -,219 ,185 6 ,195 ,109 ,111 ,157 ,157 ,845 -,021 ,262 7 ,793 -,016 ,116 ,318 ,144 ,269 ,129 ,143 8 ,628 ,259 ,431 -,072 ,137 ,063 ,386 ,112 9 ,835 ,080 ,078 ,239 ,209 ,117 ,285 ,037 11 -,026 ,334 ,117 ,693 -,215 ,076 ,338 ,048 12 ,331 ,561 -,048 ,383 -,262 ,112 ,287 ,092 13 ,363 ,380 ,336 ,541 ,036 ,113 ,082 -,121 14 ,063 ,842 ,185 ,241 ,242 ,034 ,017 ,100 15 ,078 ,260 ,251 -,014 -,014 ,036 ,187 ,834 16 ,321 ,265 ,561 ,370 ,107 ,145 -,046 ,267 17 ,428 ,456 ,371 ,161 ,118 ,257 ,348 ,121 18 ,272 -,003 ,002 ,150 ,166 ,188 -,110 ,833 19 ,105 ,197 -,120 ,471 ,370 ,382 ,356 ,377 20 ,193 ,083 ,298 ,677 ,327 ,208 -,302 ,202 21 ,330 ,309 ,204 ,511 ,542 ,064 -,012 ,166 22 ,105 ,022 ,053 -,109 ,801 ,355 -,094 ,200 23 ,182 -,001 ,148 ,016 ,640 ,663 -,086 ,005 24 ,272 -,080 ,096 ,174 ,847 -,048 ,012 -,033 25 ,060 ,055 ,939 ,022 ,122 ,058 ,053 ,068 26 ,007 -,059 ,906 ,252 ,014 ,055 ,084 ,070 27 ,109 ,154 ,140 ,150 -,081 -,086 ,878 ,037 28 ,239 -,168 ,501 ,619 ,175 -,042 ,255 -,004 Extraction Method: Principal Component Analysis.
41
Component Matrix Attitudea
Component 1 1 0,888 2 0,853 3 0,834 4 0,828 5 0,621
