The
transfer of
Sticky
Knowledge.
A research on the use of Legacy Mentors as a process to transfer knowledge from senior nurses to younger generations; moderated by the arduousness of therelationship
THE TRANSFER OF STICKY KNOWLEDGE
A research on the use of Legacy Mentors as a process to transfer knowledge from senior nurses to younger generations; moderated by the arduousness of the relationship.
by
FLEUR VERSTEGEN
Rijksuniversiteit Groningen Faculteit Economie en Bedrijfskunde
Master Thesis HRM June 2013 Hereweg 19 9725 AA Groningen (06) 23006967 f.f.verstegen@student.bdk.rug.nl studentnumber 1861972 Supervisor: drs. M. Fennis-Bregman
ABSTRACT
Companies around the globe are facing issues associated with an ageing workforce. Healthcare organizations will face the problem of double aging because the demand for care increases as well. In addition, a large number of nurses are planning to retire. Without a strategic plan to evaluate the pending retirement of their nursing workforce and transfer of the knowledge they possess, organizations may find themselves vulnerable to increased errors, loss of efficiency, and a decline in patient care. The purpose of this study is to provide decision makers in healthcare organization a helpful guideline regarding the suitability of one specific knowledge transfer process. This research has tested whether the knowledge of a workforce that may be close to retirement can be effectively transferred to less experienced generations, with the use of legacy mentoring. It was expected that this positive influence of legacy mentors on knowledge transfer was weaker when there is an arduous relationship between source and recipient.
To test the hypothesis, a field study has been conducted. Within this field study, 78 nurses working in Dutch hospitals have been asked to fill in a questionnaire survey. Although the direct positive relationship between the use of legacy mentors and knowledge transfer could be found, the results did not show a moderating effect of an arduous relationship between mentor and protégé.
Table of Content
1. INTRODUCTION ... 4
2. THEORY ... 5
2.1 Knowledge transfer ... 5
2.2 Mentoring ... 6
2.3 Knowledge transfer and mentoring ... 6
2.4 Knowledge transfer Barriers... 7
3. METHOD ... 9
3.1 Procedure & Respondents ... 9
3.2 Measurement instruments ... 11 3.3 Statistical analyses ... 12 4. RESULTS ... 13 4.1 Descriptive statistics ... 13 4.2 Hypothesis tests ... 14 4.3 Extra analyses ... 16
5. DISCUSSION & CONCLUSION ... 17
5.1 Interpretation of the findings ... 18
5.2 Theoretical Implications ... 18
5.3 Limitations... 19
5.4 Future Research ... 20
5.5 Practical Implications ... 22
REFERENCES ... 24
APPENDIX A: Questionnaire Departments with LMs ... 27
1. INTRODUCTION
Companies around the globe are facing issues associated with an ageing workforce. Healthcare organizations will face the problem of double aging because the demand for care increases as well. In addition, a large number of nurses are planning to retire. In fact, according to a Nursing Management Survey from January 2010, 49% of nurses plan to retire by 2020. These Older workers have knowledge and experience that is valuable, and “companies cannot afford to lose the wisdom and expertise of talented, experienced, knowledgeable leaders and employees, because the 78 million baby boomers cannot adequately be replaced by the 40 million people of Generation X” (Bower & Sadler, p. 45).
Although the threat that a significant number of baby-boomers will take the unique skills, knowledge, experience and relationships that they embody with them as they walk out the door is widely acknowledged in the business literature (Malone, 2002, p. 112; Streb and Voelpel, 2009, p. 305; Streb et al., 2008, p. 2; Dychtwald et al., 2004, p. 50; Burke and Ng, 2006, p. 88), in a healthcare setting it is only recently becoming a topic of discussion (Elwyn, Kowalczuk & Taubert, 2007). Without a strategic plan for hospitals to deal with the pending retirement of their nursing workforce and transfer of the knowledge they possess, organizations may find themselves vulnerable to nurse shortages, high replacement costs, increased errors, loss of efficiency, and deterioration in patient care (Sherman, 2008). An important question related to this issue, that will be the main focus of this research, is how the knowledge of experienced nurses close to retirement can be effectively transferred to less experienced generations to insure that what they know will be available for future use.
Given the heterogeneity in researched processes to transfer knowledge with varying degrees of complexity, decision makers in healthcare organizations require helpful guidelines regarding the suitability of processes within a particular healthcare setting. One example of an implementation program from the nursing industry included a focus on using mentoring to foster knowledge transfer (Clauson et al., 2011). The goal of this project was building upon the insight and knowledge of senior nurses before they leave the profession, and supporting the transfer of nursing knowledge by making them ‘Legacy Mentors’ (LMs). This research will build upon the insights of the Legacy Mentor Project (LMP) and will extend the existing research by doing a field research in which the focus is on the effect of the current use of legacy mentors on knowledge transfer.
the arduous relationship between source and recipient. The intention of this study is to make a contribution to the literature of knowledge management in healthcare settings by investigating if mentoring and an arduous relationship (between source and recipient) interact to determine the level of knowledge transfer.
2. THEORY 2.1 Knowledge transfer
Knowledge transfer occurs when the source communicates a part of his knowledge verbally or in a codified form, and the receiver absorbs this knowledge even completely (Hubig, Jonen & Lingna, 2008). The concepts of knowledge transfer, knowledge spillover, and knowledge diffusion are frequently used interchangeably in the literature, but they are not identical. According to Hubig et al. (2008) knowledge transfer is different to the terms of knowledge diffusion and knowledge spillover, because here the receiver is known and transfer activity is wanted. Szulanski (1996) uses the word 'transfer' rather than 'diffusion' to emphasize that the movement of knowledge within the organization is a clear experience, not a gradual process of diffusion, and depends on the characteristics of everyone involved.
Prior research positions the knowledge transfer phenomenon on different levels of analysis (Spraggon & Bodolica, 2012). An extensive amount of research has been done at the inter-regional, inter-industry, and inter-organizational levels rather than intra-firm level (Spraggon & Bodolica, 2012). This research seeks to contribute to the literature by concentrating on knowledge transfer processes that take place between individuals within the same organization. And more specific on the intra-firm level of knowledge transfer in healthcare organizations. The transfer of knowledge in healthcare organizations involves capturing and distributing the knowledge of experienced nurses to insure that what they know will be available for future use (Sherman, 2008).
documentation. This research seeks to contribute to the literature by concentrating on direct knowledge transfer methods that take place between individuals within the same organization. Knowledge transfer is defined here as the provision or receipt of task information, know-how, and feedback regarding a product or procedure among members (Cummings, 2001).
2.2 Mentoring
As mentioned earlier, mentoring is a practice that supports direct knowledge transfer. Delong (2004) suspect that mentoring and coaching are the most effective direct practices to transfer critical implicit and tacit work related knowledge. The reason behind this line of reasoning is that mentoring supports the sharing of the broadest set of knowledge, from explicit technical skills, cultural values, operational or managerial skills to career development advice.
A closer look at the concept of mentoring reveals some troubling issues. Jacobi (1991) mentions two major concerns related to the concept of mentoring; the absence of a widely accepted operational definition of mentoring which results in a second concern, which is a continued lack of clarity about the antecedents, outcomes, characteristics, and mediators of mentoring relationships despite a growing body of empirical research. This supports Merriam’s (1983) proposition that mentoring has a different meaning to business people, psychologists, and people in academic settings.
Furthermore, the concept of mentoring is frequently used interchangeably with the concept of coaching, but they are not identical. According to Delong (2004), mentoring always involves coaching, but coaching does not necessarily involve a mentoring relationship. In the context of knowledge transfer, mentoring is a broader concept that implies more interest in the receiver’s long-term development.
2.3 Knowledge transfer and mentoring
decision makers require helpful guidelines regarding the suitability of the chosen process for a given corporate setting (Spraggon & Bodolica, 2012).
One example of an implementation program that does stand out is a pilot knowledge management program from the nursing industry that reported an ageing workforce best practice including a focus on using mentoring for knowledge transfer (Clauson et al., 2011). The goal of this project was building upon the insight and knowledge of senior nurses before they leave the profession, and supporting the transfer of nursing knowledge by making them ‘Legacy Mentors’ (LMs). LMs are nurses aged 55 or older who have a wealth of knowledge and experience to share with other nurses. The Legacy Mentor Project (LMP) was a unique “project within a project” in which the participants were selected on the basis of their interest in the project. The project consisted of several workshops in which the success was really determined by the fact that the participants choose projects in which they had a keen interest. Findings of this LMP showed that the translation of expertise by re-energized nurses has the potential to increase retention of the most experienced nurses while also enhancing practice learning environments. This paper will extend the existing research by doing a field research in which the focus is on the effect of the current, general use of Legacy Mentors on knowledge transfer.
Older nurses require opportunities to share what they have learned throughout their career (Bower and Sadler, 2009). The specific goal of this mentorship is to transfer knowledge from senior nurses to students, novice nurses, new hires, and/or experienced staff (Clauson et al., 2011). Therefore, the following hypothesis is formulated:
Hypothesis 1: The use of Legacy Mentors is positively related to knowledge transfer.
2.4 Knowledge transfer Barriers
Szulanski called the less known set of reasons knowledge barriers. Examples of knowledge barriers are the recipient’s level of knowledge prior to the transfer, recipient's lack of absorptive capacity, causal ambiguity, and an arduous relationship between the source and the recipient. The study findings showed that knowledge barriers to transfer have a larger effect on the stickiness of knowledge than motivational barriers (Szulanski, 2003). Therefore, the focus in this research will be on knowledge barriers, more specific on the arduous relationship between source and recipient, in which an arduous relationship can be defined as a relationship that misses ease of communication and intimacy of the relationship (Szulanski, 2000).
2.5 The moderating effect of an arduous relationship on knowledge transfer
The second hypothesis is based on the work of multiple researches suggesting that knowledge barriers to transfer have a large effect on the stickiness of knowledge. Especially when the knowledge that is transferred consist tacit components, it may require numerous individual exchanges (Nonaka, 1994). Tacit knowledge is hard to communicate and is deeply rooted in action, involvement and commitment within a specific context (Nonaka, 1994, p. 16). As Polanyi (1966) noted, individuals know more than they can explain, since individuals possess tacit knowledge that is non-verbalized, intuitive, and unarticulated. Research has shown that articulable knowledge is more easily transferable than less articulable Knowledge (Cummings & Teng, 2003). In addition, the success of knowledge transfer depends to some extent on the ease of communication (Arrow, 1974) and on the 'intimacy' of the over-all relationship between the source and the recipient (Marsden, 1990). An arduous (i.e., laborious and distant) relationship might create additional hardship in the transfer (Szulanski, 1996). Therefore, the following hypothesis is formulated:
Hypothesis 2: The positive influence of Legacy Mentors on knowledge transfer is weaker when there is arduous relationship between source and recipient.
H2
-
H1 +
FIGURE 1:
Conceptual Model;Arduous relationship as a moderator on the relationship between the use of Legacy Mentors and knowledge transfer.
The intention of this study is to make a contribution to the literature of knowledge management in healthcare settings by investigating if mentoring and an arduous relationship (between source and recipient) interact to determine the level of knowledge transfer. Central to this study is the following research question:
“How do the use of Legacy Mentors and the existence of an arduous relationship (between
source and recipient) interact to determine the level of knowledge transfer?”
3. METHOD 3.1 Procedure & Respondents
The data for the present study is collected via paper-and-pencil questionnaires and online questionnaires. All participants were nurses working in Dutch hospitals. Two research departments of two different Dutch hospitals were contacted by phone and email, to arouse their interest to participate in this study. The rest of the participants are working at 4 other Dutch hospitals, and are contacted with the help of social contacts (via Facebook and email), and hospital visits. The content and purpose of this study is explained to the organizations but was not communicated to the participants of this research, to avert undesirable effects.
In advance of this survey it is determined by phone, email or visiting, whether a department uses Legacy Mentors. In order to test if the use of LMs leads to more knowledge transfer, the level of knowledge transfer is compared between departments that make use of
Use of Legacy Mentors
Arduous relationship between source and recipient
LMs and departments that do not. Approximately an equal number of participants were: LMs - and nurses older than 55 years who are not a LM, nurses younger than 55 years that have a LM – and nurses younger than 55 years who do not have a LM (see table 1). In this way, it was possible to obtain a balanced perspective on each transfer. Furthermore, by distinguishing between these four types of respondents, it was possible to perform extra analyses in order to test whether any results from this study could be explained by differences in mean scores on (in)dependent variables between these four types of participants.
Two different versions of the questionnaire are made; one for the departments that do make use of LMs (see appendix A), and one for departments that do not (see appendix B). In this way it was possible to specify the terms ‘source’ and ‘recipient’. In the questionnaire for departments that do make use of LMs, the term ‘source’ is replaced by ‘mentor’. In the questionnaire for departments that do not make us of LMs, the term ‘source’ is replaced by ‘baby boomers’ and the term ‘recipient’ is replaced by ‘younger generations’. The questionnaires included questions related to tenure, age, and gender of the respondent. A concise introductory text explicitly guaranteed that the responses of the participants would be treated with the highest confidentiality.
The departments that do make use of LMs, made use of both paper-and-pencil questionnaires and online questionnaires. Short questionnaires were sent to all the 55 team members working within the two different departments that make of use of LMs across the two Dutch hospitals that had agreed to participate in this study. To obtain a balanced perspective on each transfer, one questionnaire has been sent to the LM and one to their protégés. Thirty-five questionnaires were fully completed with no missing data (response rate is 64%), in which fourteen Legacy Mentors and twenty-one protégés.
Meanwhile, the online survey link was forwarded to my social relationship, via e-email, and Facebook, who are either working as a nurse or have connections with Dutch nurses. In addition, during visits to the hospital, nurses were asked to fill in the paper-and-pencil questionnaire. In this way, it was possible to add the questionnaires with some quantitative data. In total, this resulted in forty-three fully completed questionnaires with no missing data from departments that do not make use of LMs.
Table 1 provides an overview of the characteristics of the two different samples and shows that the percentage of female, the mean age, and mean tenure differed between the nurses <55 years and protégés. Further analysis showed that these background variables did not correlate significantly with one of the independent variables or dependent variable of this study (see table 2). For this reason gender, age, and tenure is not controlled in further analyses.
Table
1
: Overview of the Composition of the Data Sets at the two types of
departments
n Gender (%female) Age M SD Mean TenureOrg Team Position
Departments that make use of LMs Legacy Mentors 14 92,9 57,4286 2,59331 26,7143 17,7857 12,7143 Protégés 21 90,5 26,4762 2,94311 5,7143 4,8095 4,3810 Departments that don’t make use of LMs Nurses < 55 years 28 92,9 33,3929 8,18463 9,1786 6,6429 6,1786 Nurses > 55 years 15 93,3 57,9333 2,37447 26,4667 17,2667 14,5333 3.2 Measurement instruments
To test the above hypotheses, data was gathered on perceptions of knowledge transfer and the arduousness of the relationship between source and recipient. The paper-and-pencil survey and online survey conducted the same questionnaire, which was originally used in English and subsequently translated into Dutch by author. Cross-validation was used to make sure that the translated questionnaire is not too much deviated from the original versions. The questionnaire started with questions about demographical variables. Furthermore, the questionnaire contained five self-report scales of knowledge transfer, and three questions related to the arduousness of the relationship between source and recipient.
Use of Legacy Mentors In advance of this survey it is determined by phone, email or
visiting, whether a department uses LMs, using the definition of Clauson et al. (2011). In this study the use of legacy is a dichotomous independent variable with the values: 0 = no use of LMs, 1 = use of LMs. In this way it is possible to compare the level of knowledge transfer between departments that do make use of LMs with departments that do not.
Internal knowledge transfer. The level of internal knowledge transfer was measured
which mentor and protégé or baby boomers and younger generations transfer: (a) objectives and responsibilities; (b) specific requirements; (c) written procedures and the practical knowledge to use them; (d) superior practices for carrying out tasks; and (e) clear recommendations and tips to improve the performance of the tasks, ranging from 1 (never) to 7 (to a great extent). An example of an item is: “Indicate with what frequency practices that
have proven to be superior in their application are transferred between source and recipient.” The internal consistency for the present study (α = 0.91), indicates an acceptable
level. In the study of Cummings (2001) the internal consistency was considerably lower (α = .78).1
Arduous relationship. The arduousness of the relationship between source and
recipient was measured using a three-item scale developed by Szulanski (1996). Scale items related to the communication and collaboration between source and recipient. Participants are requested to indicate on a four-point-ordinal-scale the ease of communication and collaboration between the source and recipient, ranging from 1 (very easy) to 4 (very difficult. An example of such a question is: “Collaboration between source and recipient: 1-Is sought
actively by source. 2-Is well received but not sought by source. 3- Is preferably avoided by source. 4- Occurs only if source has no choice.” The internal consistency was studied (α=
0.72), and indicates an acceptable level after deleting one of the three items. In the study of Szulanski (1996) the internal consistency was approximately equal (α= 0.71), but was measured using all three items.
Control variables. Previous research has indicated that there may be important
individual-level factors that have an influence on knowledge sharing behavior, such as gender (Bassam, Mesbah & Jafar, 2009 and Mäkelä, Andersson & Seppälä, 2012), age, and tenure (MacCurtain et al., 2010). Therefore, it is considered to be necessary to measure these possible differences between the respondents. These background variables were measured as a section of the questionnaire by using single item questions.
3.3 Statistical analyses
In order to test the conceptual model and stated hypotheses, the following procedures have been followed:
In order to test whether the variables, which are assessed by using different items, can be computed, a correlation analysis has been conducted and the Cronbach’s Alpha of the
moderator (arduous relationship) and dependent variable (knowledge transfer) are calculated. Only two of the three items of the moderator correlated significant (α= 0.72). Therefore, one item had to be deleted. All of the items that measured knowledge transfer correlated significant and had α >0.6 (α= 0.91), for this reason no item had to be deleted.
The second step consisted in standardizing the variables. Since only one of the independent variables (arduous relationship) is a continuous variable is, only this variable has to be standardized.
Third, a hierarchical regression analysis has been conducted in order to test if the expected positive influence of Legacy Mentors on knowledge transfer is weaker when there is arduous relationship between source and recipient. This regression consisted of 2 steps. In the first step the unstandardized independent variable (knowledge transfer) and the standardized moderator (arduous relationship) are added to control for the effect of the independent variable and the effect of the moderator. In step 2, the interaction term (knowledge transfer*arduous relationship) was added in order to investigate if the proposed conditional direct effect exists.
Finally, an extra analysis has been conducted to test whether the expected positive relation between the use of LMs and knowledge transfer can be explained by the high level of perceived knowledge transfer of LMs. In order to test this, an independent sample t-test has been conducted. Within this t-test, the mean level of perceived knowledge transfer of LMs is compared to the mean level of perceived knowledge transfer by their protégés.
4. RESULTS
In this section, we will take a look at the results from the hierarchical regression analysis which is conducted in order to test how the use of LMs and an arduous relationship (between source and recipient) interact to determine the level of knowledge transfer. First, the correlation table will be presented, followed by the results from the hierarchical regression analysis and extra analyses.
4.1 Descriptive statistics
knowledge transfer, and the independent variables; arduous relationship, and the use of Legacy Mentors (LMs).
As expected, the use of Legacy Mentors is indeed positively correlated to knowledge transfer (r = .50, p < .01). In addition, a negative significant correlation was found between an
arduous relationship and knowledge transfer (r = -.59, p < .01). Another negative significant
correlation was found between an arduous relationship and the use of LMs (r = -.22, p < .05). With regard to the control variables, it can be stated that no significant correlations were found between gender (r = -.08, p = n.s.), age (r = 0.06, p = n.s.), organizational tenure (r = .06, p = n.s.), team tenure (r = .06, p = n.s.), positional tenure (r = -.05, p = n.s.) and knowledge transfer. Therefore, it can be concluded that there is no need to take these background variables into account as control variables in further analyses. On the other hand, from these correlation table it is clear that there are positive significant correlations between organizational tenure (r = .86, p < .01), team tenure (r = .71, p < .01), positional tenure (r = -.55, p < .01) and age.
TABLE 2: Explorative data (means, SD and Pearson correlations)
Mean SD 1 2 3 4 5 6 7 1. Gender 1,07 0,27 2. Age 40,56 14,53 -0,06 3. Org.Tenure 14,72 11,26 -0,07 0,86** 4. Team Tenure 10,19 8,29 -0.09 0,71** 0,81** 5. Pos.Tenure 8,47 7,56 -0,11 0,55** 0,55** 0,68** 6. Use of LMs 0,45 0,50 0,03 -0,11 0,05 -0,02 -0,09 7. Arduous relationship 1,21 0,37 0,03 -0,10 -0,12 -0,07 -0,13 -0,22* 8. Knowledge transfer 5,33 0,85 -0,08 0,06 0,06 -0,05 0,04 0,50** -0,59** n= 78, *p<.05, **p<.01 4.2 Hypothesis tests
Step 1 tested the effect of the use of LMs and the arduousness of the relationship (between source and recipient) on knowledge transfer. Table 3 shows that these variables explained 0.47 variance of the criterion knowledge transfer. As expected, both variables predicted knowledge transfer significantly (R²= 0.48, β = 0.39; -0.52, F = 35.23, p < 0.01). Hypothesis 1 stated that the use of Legacy Mentors is positively related to knowledge transfer.
Therefore, hypothesis 1 was supported in the present study.
TABLE 3: Hierarchical regression analysis with the use of Legacy Mentors and the arduousness of the relationship as predictors and knowledge transfer as criterion
Knowledge transfer
Independent variables β SE F R² ∆R² Sig.
Step 1 Use of LMs 0,39 0,08 35,23 0,48 0,47 0,00**
Arduous relationship -0,52 0,08 35,23 0,48 0,47 0,00**
Step 2 Use of LMs * Arduous relationship 0,04 0,09 23,29 0,49 0,47 0,69
n = 78, *p < .05, **p < .01
During step 2, the standardized moderator (use of LM’s*arduous relationship) is added. This step was used to test hypothesis 2, which stated that the positive influence of LMs on knowledge transfer is weaker when there is arduous relationship between source and recipient. As table 3 shows, the interaction effect did not have a significant contribution in predicting knowledge transfer (R² = 0.49, F = 23.29, p = n.s.). In Figure 2, it can also be seen that, this hypothesis cannot be confirmed. Therefore, hypothesis 2 was not supported in the
FIGURE 2:
Arduous relationship as a moderator on the relationship between the use of Legacy Mentors and knowledge transfer.
4.3 Extra analyses
With the use of self-reporting measures, validity problems may arise. For example, it is possible that LMs exaggerated the level of knowledge transfer in order to make their level of knowledge transfer seem higher. To test whether the founded positive relation between the use of LMs and knowledge transfer can be explained by the high level of perceived knowledge transfer of LMs, an independent sample t-test has been conducted. With this t-test, the mean level of perceived knowledge transfer of LMs is compared to the mean level of perceived knowledge transfer by their protégés. The results of this test do show a difference between these two groups in the level of perceived knowledge transfer (see table 4), although the difference between LMs and protégés in perceived knowledge transfer is not significant (mean difference = .33, p = n.s.).
TABLE 4: Independent Samples T-test, comparing the level of perceived knowledge transfer between LMs and their protégés.
Knowledge transfer
N mean SD SE t Sig
Protége 21 5,66 0,54 0,12 -1,44 0,16
LM-role 14 5,99 0,73 0,19
Although the expected interaction effect between the use of Legacy Mentors and the existence of an arduous relationship (between source and recipient) could not be found in the present study, there are indications of partial mediation. It might be that the arduousness of the relationship (between source and recipient) is not a moderator, but could in fact mediate the relationship between the independent variable (use of LMs) and dependent variable (knowledge transfer). As expected, the use of LMs is indeed positively correlated to knowledge transfer (r = .50, p < .01). In addition, a negative significant correlation was found between an arduous relationship and knowledge transfer (r = -.59, p < .01). Another negative, significant correlation was found between an arduous relationship and the use of LMs (r = -.22, p < .05). To find out if there might be (partial) mediation, a Sobel test has been conducted. The results of this Sobel test showed however no significant (partial) mediation effect (p = 0.06).
5. DISCUSSION & CONCLUSION
The analyses generally support the findings from prior research that the use of mentoring results in more (perceived) knowledge transfer (e.g., Clauson et al., 2011; Delong, 2004; Spraggon & Bodolica, 2012; Pollac, 2012; Eby, Rhodes, & Allen, 2007; Malone, 2002; Streb and Voelpel, 2009; Streb et al., 2008; Dychtwald et al., 2004; Burke and Ng, 2006). The amount of knowledge transfer was highest when older, experienced nurses were used as Legacy Mentors and lower when these baby boomers were not used as Legacy Mentors. The pattern of results with regard to an arduous relationship turned out to be more complex than expected. In line to what is suggested in the knowledge literature (Szulanski, 2003; Elwyn et al., 2007), a negative correlation between an arduous relationship and knowledge transfer was found. Furthermore, a negative correlation between an arduous relationship and the use of LMs was found. From insights gained from the knowledge transfer literature (Szulanski, 2003; Elwyn et al., 2007), it was hypothesized that when an arduous relationship between the source and recipient existed, the positive influence of LMs on knowledge transfer would be weaker, but the results do not support this.
5.1 Interpretation of the findings
An initial potential explanation for the negative correlation between the existence of an arduous relationship and the use of LMs may lie in the fact that mentoring is difficult to establish and mentors and mentees preference similarity (Moreland & Levine, 2006). In this way, most of the time combinations between mentor and recipient are only made when these parties can communicate and collaborate easily with each other.
A second, related explanation might be that participation as a mentor could be influenced to some degree by personality (Career Development International. 2006). The results of earlier research indicated that extroversion, conscientiousness, and openness to experience were positively correlated with participation as a mentor (Career Development International. 2006). In addition, previous research has suggested that people who score high on extraversion, conscientiousness, and openness to experience are more likely to build developmental (high-strength) relationships (Dougherty, Cheung & Florea, 2008). So a combination of both findings may explain the unexpected negative correlation between the existence of an arduous relationship and the use of LMs.
A potential explanation for the absence of finding an interaction effect between an arduous relationship and the use of Legacy Mentors may lie in the fact that the Legacy Mentors had to generalize their answers about the arduousness of their relationship with their protégés instead of answering the questions for each protégés individually. It is possible that for this reason less extreme results are found in the arduousness of relationships.
As is evident from the discussion above, a lot of work still needs to be done to fully understand the precise relationships among the use of LMs, the relationships between mentor and recipient, and knowledge transfer. Some interesting directions for future research will be discussed in the next section. Despite the fact that the findings of the present study do indicate a need for more research, they offer likewise several meaningful insights.
5.2 Theoretical Implications
Furthermore, the findings from the extra analyses reveal that a small proportion of the variance in perceived knowledge transfer were due to differences in perceived knowledge transfer between Legacy Mentors and nurses with a Legacy Mentor. The finding that LMs perceived slightly more knowledge transfer than their protégés, adds to existing knowledge (e.g., Clauson et al., 2011; Palmer & Schoorman, 2011) by providing a deeper insight into the relatively unknown processes of how different parties in the knowledge transfer process rates the level of knowledge transfer. The fact that the answers of the Legacy Mentors explained a small portion of the positive relationship between the use of LMs and knowledge transfer, implies that it would be beneficial to broaden the traditional focus in the knowledge literature about the use of mentoring, which tends not to make a distinction between the experienced level of knowledge transfer between different parties within this knowledge transfer process (e.g., Clauson et al., 2011; Palmer & Schoorman, 2011).
Thirdly, the analysis found a negative relationship between an arduous relationship and the use of LMs. The relationship between the mentor and recipient tended to be perceived as less arduous than was generally the case between baby boomers and younger generations. These results contribute to theories about mentoring by indicating that the use of Legacy Mentors correlates with less arduous relationships between older and younger generations in a healthcare setting.
Fouthly, the results support Szulanski’s (1996) contentionthat an arduous relationship, between source and recipient, has a negative effect on knowledge transfer. Moreover, the results of this study extend current knowledge by suggesting that collaboration is a better determinant of an arduous relationship than communication. Since, the results showed a stronger internal consistency when only the last two items (which measured collaboration between source and recipient) of this variable were computed.
5.3 Limitations
group of people at different points in time (Field, 2005), and include a third party of the transfer to provide firmer evidence for these findings.
Another potential weakness of the present study is the use of a three-item measurement for the arduous relationship variable. In line with another study involving knowledge transfer barriers (e.g., Szulanski, 2006), a three-item measurement was opted because the demands on the respondents were already quite high. However, given the fact that the items were adopted from an existing scale hat have shown good reliability and validity in previous research (Szulanski, 2006), it is not expected that the use of this three-item measurement would invalidate the conclusion. Nevertheless, future research might benefit from the design of a scale including more items to measure this variable.
Thirdly, in the sample of this study the number of departments that make use of LMs is much lower as compared to the number of departments that do not make use of LMs. So in terms of number of departments, the sample of departments that make use of LMs was more homogeneous. This homogeneity raises questions about the generalizability of our findings to other settings. Conversely, this homogeneity could also be seen as an advantage, because it is very representative for the healthcare industry, since there are very few departments that do make use of LMs.
Fourthly, the present study is conducted in Dutch hospitals, the research results may be affected to some extent by the culture of Netherland. Therefore, the generalizability of these findings to other nations is still in question.
Fifthly, the present study only conducted correlational research, and did not take into consideration cause-effect relations. For instance, the cause and effect relation between an arduous relationship and the use of Legacy Mentors has not been taken into consideration. Therefore, future research could conduct experimental research to see its effect on another.
Finally, it was not possible to control the determination whether a person indeed have a LM or fulfill the role of LM, when they used the online questionnaire. As was explained in the method section, it was determined by phone, email or visiting, whether a department makes use of LMs. In the online-setting it was more difficult to make sure that the respondents had the right vision of how the present study defines LMs.
5.4 Future Research
whether there exist partial mediation between the use of Legacy Mentors, an arduous relationship, and knowledge transfer, instead of the expected interaction effect between these variables. The results of the Sobel test in the present study showed an almost significant (partial) mediation effect (p = .06). Future research could examine with a new set of respondents whether a significant partial mediation effect can be found, when all the limitations of the present study are taken into account.
Since only partially mediation can be found in future research, it suggests that other variables must also influence this relationship. Future research might therefore consider the role of other mediating variables as well. Three promising variables to consider derive from model of Szulanski (2003). In this study only the effect of one of the four knowledge barriers, described by Szulanski (2003), on knowledge transfer is tested. Future research could therefore test whether the remaining three knowledge barriers (the recipient’s level of knowledge prior to the transfer, recipient's lack of absorptive capacity, and causal ambiguity) (partially) mediate the relationship between the use of LMs and knowledge transfer.
Earlier research examined the role of trust as a moderator of the relationship between mentoring and knowledge transfer (Palmer & Schoorman, 2011). According to Palmer & Schoorman (2011), the relationship between mentoring and knowledge transfer is stronger when there is low trust, than when the level of trust is high. One implication that derived from that study is that trust is most important when mentoring support is low. As this suggestion illustrates, the inclusion of moderators such as trust between the mentor and protégé might help to explain when and why the use of Legacy Mentors leads to more knowledge transfer.
Building on this, it might be interesting to investigate if the use of LMs also works as a strategy to retain older generations, close to retirement because they may felt valued as a result of providing something of value to others (Clauson et al., 2011). If this is the case, this will help the healthcare industry in dealing with the capacity problem at least on the short-term. In relation to the research model of the present study, the retention of these older nurses can result in an extension of the period in which they are working as Legacy Mentors, which in turn may lead to more knowledge transfer, or the transfer of more complex knowledge.
transfer. For instance, when future research indicates a negative correlation between extroversion and an arduous relationship, organizations could search for extrovert employees in order to decrease the arduousness of relationships, which in turn has a positive effect on knowledge transfer.
Another suggestion is that, as a further refinement to this study, one could investigate the role of the task environment (e.g., Thompson, 1967). It seems logically that in some nurse departments, it is less likely to transfer knowledge about best practices than in other nurse departments with more standardized tasks. Future research is needed to establish whether the use of LMs would indeed have the desired effects in multiple nursing departments. Therefore, it is reasonable that the results indicate a need for further research into knowledge transfer in hospitals and hopefully this study provides a good starting point and encouragement for undertaking such examinations.
A final suggestion, as a further refinement to this study as well, is to investigate how the mentor-relationship can be shaped most effectively in practice. The results from this study show that the use of LMs in general, leads to more knowledge transfer. In order to able to be more specific about the shape of this relationship, one could investigate which frequency of interaction, between mentor and protégés, leads to the most knowledge transfer.
5.5 Practical Implications
Before the Legacy Mentoring programs can be implemented, the organizations should be prepared for knowledge transfers. First, organizations need to know which older workers will soon be leaving the organization and which valuable knowledge may disappear. In reaction to this, organizations need to plan to whom, and in what time frame the implicit knowledge should be transferred.
Secondly, younger workers should be open to the knowledge, capable of adapting the knowledge, and adopt an interested attitude. For older workers, communication and dealing with young people are important issues. Also the older workers should be willing and like to transfer knowledge before they leave. In addition, older workers should communicate strong enough to communicate understandable even if it concerns complex knowledge.
Thirdly, the progress of the knowledge transfer process should be followed. When necessary, employees should get feedback or extra stimulation to transfer knowledge.
Fourthly, the management team should be taken responsible for the strategy and implementation of this program. It is important that they will manage and facilitate the process top down; older employees should be stimulated to participate in the project and the value of their knowledge should be emphasized. Leaders should ensure that knowledge transfer actually takes place in their team. And the HR-management should be involved when, despite the good intentions, no knowledge transfer takes place.
REFERENCES
Arrow, K. J. (1969). Classification notes on the production and transmission of technical knowledge, American Economic Review; Papers and Proceedings, 52, 29-35. Bassam, H., Mesbah, A., & Jafar, A. (2009). A model of gender differences in knowledge
sharing attitude in group projects. Northeast Region Decision Sciences Institute. 6, 185-190
Bower, F.L., Sadler, W.A. (2009). Why retire? Career strategies for third age nurses. Sigma Theta Tau International Honor Society of Nursing.
Burke, R. & Ng, E. (2006). The changing nature of work and organizations: implications for human resource management. Human Resource Management Review, 16, 86–94 Career Development International (2006). Personality predictors of participation as a mentor
Career Development International, 11, 321-333.
Clauson, M., Wejr, P., Frost, C., McRea, D., Straight, H. (2011). Legacy mentors: Translating the wisdom of our senior nurses. Nurse education in practice, 11, 153-158
Cummings, J.N. (2001). Work group and knowledge sharing in a global organization.
Management Science, 50(3), 352-364.
Cummings, J.L. & Teng, B. (2003). Transferring R&D Knowledge: The Key Factors Affecting Knowledge Transfer Success. Journal of Engineering and Technology
Management, 20, 39–68.
Delong, D.W. (2004). Lost Knowledge: Confronting the Threat of an Aging Workforce. New
York: Oxford University Press; 11-12.
Dougherty T. W., Cheung Y. H., Florea L. (2008). The role of personality in employee developmental networks. Journal of Managerial Psychology, 23, 653-669
Dychtwald, K., Erickson, T. & Morison, B. (2004). It's time to retire retirement. Harvard
Business Review, 48–57
Field, A. (2005). Discovering statistics using spss. London: SAGE publications.
Eby, L. T., Rhodes, J. E., & Allen, T. D. (2007). Definition and evolution of mentoring. In
The Blackwell handbook of mentoring: A multiple perspectives approach. Malden,
MA: Blackwell
Knowledge Transfer in Universities: A German Perspective. ICFAI Journal of
Knowledge Management, 6, 32-46
Jacobi, M. (1991). Mentoring and Undergraduate Academic Success:A Literature Review.
University of California-Los Angeles, 61, 505-532.
Leonard-Barton, D. (1990). The intraorganizational environment: Point-to-point versus diffusion. In F. Williams and D. V. Gibson (eds.), Technology Transfer: A Communication Perspective. Sage, Lon-don, 43-62.
MacCurtain, S., Flood, P., Ramamoorthy, N., West, M., Dawson, J. (2010). The Top
Management Team, Reflexivity, Knowledge Sharing and New Product Performance: A Study of the Irish Software Industry. Creativity & Innovation Management. 19, 219-232
Mäkelä, K., Andersson, U., & Seppälä, T. (2012). Interpersonal similarity and knowledge sharing within multinational organizations. International Business Review. 3:439-451 Malone, D. (2002) Knowledge management: a model for organizational learning.
International Journal of Accounting Information Systems, 3:111–123
Marsden, P. V. (1990). Network data and measure-ment. Annual Review of Sociology, 16: 435- 463.
Merriam, S. (1983). Mentors and proteges: A critical review of the literature. Adult
Education Quarterly, 33, 161-173.
Moreland, R.L. & Levine, J.M. (2006). Socialization in organizations and work groups.
Group processes & intergroup relation, 5, 185–201.
Nonaka, I. (1994). A dynamic theory of organizational knowledge creation, Organizational
Science, 5(1), 14-37.
Palmer, F. & Schoorman, M. (2011). Trust as a Moderator of the Relationship Between Mentoring and Knowledge Transfer. Journal of Leadership & Organizational Studies, 18: 334-343.
Polany, M. (1966). The logic of tacit inference. Philosophy, 4:1–18
Pollac, J. (2012) Transferring knowledge about knowledge management: Implementation of a complex organisational change programme. International Journal of Project
Management, 30: 877–886
Sherman, R.O. (2008). Lost knowledge: confronting the challenges of an aging nursing workforce. Nurse Leader, 45–56
Streb, C. & Voelpel, S. (2009) Analyzing the effectiveness of contemporary ageing workforce management: the case of Daimler AG. Organizational Dynamics, 38:305–331
Streb, C., Voelpel, S. & Leibold, M. (2008). Managing the aging workforce: status quo and implications for the advancement of theory and practice. European Management
Journal, 26: 1–10
Szulanski, G. (1996). Exploring Internal Stickiness: Impediments to the Transfer of Best Practice Within the Firm. Strategic Management Journal, 17: 27-43
Szulanski , G. (2003). Sticky knowledge: barriers to knowing in the firm. London: Sage Publications Limited; 2003.
Szulanski , G. (2000). The Process of Knowledge Transfer: A Diachronic Analysis of Stickiness. Organizational Behavior and Human Decision Processes, 82: 9-27. Teece, D. (1977). Technology transfer by multi-national corporations: The resource cost of
transfer-ring technological know-how, Economic Journal, pp. 242-261.
Thompson, J. D. (1967). Organizations in action: Social science bases of administrative
theory. New York: McGraw-Hill
APPENDIX A: Questionnaire Departments with LMs
DEEL 1
Gelieve onderstaande in te vullen:
1. Geslacht
o Man
o Vrouw
2. Leeftijd (in jaren): ……
3. Aantal jaren werkzaam in huidige organisatie: ……. 4. Aantal jaren werkzaam in huidig team: ……. 5. Aantal jaren werkzaam in huidige positie: ……. 6. Binnen mijn huidige positie:
o Vervul ik een mentor-rol o Heb ik een mentor
Dit is een: o Man o vrouw
o Anders, namelijk: …………
DEEL 2
Geef aan met welke frequentie de verschillende soorten van kennis, vaardigheden, technieken, en informatie gemiddeld worden overgedragen of uitgewisseld tussen mentor en ontvanger. (1—nooit; 7-in grote mate)
1 2 3 4 5 6 7 Nooit In grote mate
7. Huidige doelstellingen,
verantwoordelijkheden of activiteiten van de groep of sectie
1 2 3 4 5 6 7
8. Specifieke eisen van projecten 1 2 3 4 5 6 7
9. Schriftelijke procedures en praktische kennis
die nodig is om ze te initiëren
1 2 3 4 5 6 7
10. Praktijken die superieur bewezen zijn in hun
toepassing
1 2 3 4 5 6 7
11. Duidelijke aanbevelingen en trucs voor het
verbeteren van de prestaties van de processen.
DEEL 3
Omcirkel het antwoord dat het meest van toepassing is.
12. Communicatie tussen mentor en ontvanger is:
a) Zeer makkelijk b) Redelijk makkelijk c) Vrij veeleisend d) Zeer veeleisend
13. Samenwerking tussen mentor en ontvanger:
a) Wordt nagestreefd door de mentor
b) Wordt goed ontvangen maar niet nagestreefd door de mentor c) Wordt bij voorkeur vermeden door de mentor
d) Treedt alleen op indien de mentor geen keuze heeft
14. Samenwerking tussen mentor en ontvanger:
a) Wordt nagestreefd door de ontvanger
b) Wordt goed ontvangen maar niet nagestreefd door de ontvanger c) Wordt bij voorkeur vermeden door de ontvanger
APPENDIX B: Questionnaire Departments without LMs
DEEL 1
Gelieve onderstaande in te vullen:
12. Geslacht
o Man
o Vrouw
13. Leeftijd (in jaren): ……
14. Aantal jaren werkzaam in huidige organisatie: …… 15. Aantal jaren werkzaam binnen huidig team: ……. 16. Aantal jaren werkzaam in huidige positie…….
DEEL 2
Geef aan met welke frequentie de verschillende soorten van kennis, vaardigheden, technieken, en informatie gemiddeld worden overgedragen of uitgewisseld tussen de generatie babyboomers (± 55 jaar en ouder) en jongere generaties verpleegkundigen. (1—nooit; 7-in grote mate)
1 2 3 4 5 6 7 Nooit In grote mate
17. Huidige doelstellingen,
verantwoordelijkheden of activiteiten van de groep of sectie
1 2 3 4 5 6 7
18. Specifieke eisen van projecten 1 2 3 4 5 6 7
19. Schriftelijke procedures en praktische kennis
die nodig is om ze te initiëren
1 2 3 4 5 6 7
20. Praktijken die superieur bewezen zijn in hun
toepassing
1 2 3 4 5 6 7
21. Duidelijke aanbevelingen en trucs voor het
verbeteren van de prestaties van de processen.
Deel 3
Omcirkel het antwoord dat het meest van toepassing is.
11. Communicatie tussen babyboomers en jongere generaties verpleegkundigen is:
a) Zeer makkelijk b) Redelijk makkelijk c) Vrij veeleisend d) Zeer veeleisend
12. Samenwerking tussen babyboomers en jongere generaties verpleegkundigen:
a) Wordt nagestreefd door de babyboomers
b) Wordt goed ontvangen maar niet nagestreefd door de babyboomers c) Wordt bij voorkeur vermeden door de babyboomers
d) Treedt alleen op indien de babyboomers geen keuze hebben
13. Samenwerking tussen babyboomers en jongere generaties verpleegkundigen:
a) Wordt nagestreefd door de jongere generaties
b) Wordt goed ontvangen maar niet nagestreefd door de jongere generaties c) Wordt bij voorkeur vermeden door de jongere generaties
