This study extends the pygmalion effect that was generally researched in the education sector to the corporate sector. Based on the multi-dimensional pygmalion leadership, this study aims at analyzing the relationship of pygmalion leadership with LMX quality, emotional response(sense of self-efficacy, work motivation) and execution of duties. These research objectives are explained in detail as seen below.
First, this study aims at verifying the linear relationship and distinguished relationship of the pygmalion leader behavior with that of LMX quality. The pygmalion leader behavior reflects the level of anticipation of superiors, and the high level of anticipation of superiors have a high possibility to create high levels of LMX. Therefore, it is expected that the pygmalion leader behavior will have a positive effect on LMX quality.
Second is to verify the linear relationship and distinguished relationship of the pygmalion leader behavior with that of the subordinate's emotional response(sense of self-efficacy, work motivation). Recent leadership theories perceive arousing strong emotions of subordinates as an important issue. In the case of pygmalion leadership, if the subordinates' emotional response is not promoted based upon the expectations and trust of superiors and subordinates, the sustainability of leadership will fall. Thus, this study aims at defining the positive (+) relationship between the emotional response of leadership and subordinates.
Third is to verify the linear relationship and distinguished relationship of the pygmalion leader behavior with that of the subordinate's execution of duties(organization citizen behavior, innovation behavior, task performance). Because leadership is the exercising of influence to achieve an objective, it is ultimately necessary to relate pygmalion leadership to the subordinate's successful execution of duties. Hence, this study aims at approaching execution of duties from several dimensions to verify the effects of pygmalion leadership on execution of various duties.
Fourth is to verify the relationship between the subordinate's emotional response(sense of self-efficacy, work motivation) and the subordinate's execution of duties(organization citizen behavior, innovation behavior, task performance). Since long ago, studying the emotional responses that have strong effects on execution of duties have been a major point of interest for scholars studying organization behavior. Recently, there have been much research on the effectiveness as strong emotional responses and work motivation has been used as a major variable for emotional response since the past. This study aims at systematically verifying the effects of these emotional responses on the different execution of duties.
Fifth is to define the structural relationships of pygmalion leader behavior, LMX quality, emotional response and execution of duties. The structure model that stretched from pygmalion leadership, LMX quality, emotional response and to execution of duties are examined through covariance structure analysis.
In this study, previous research achievements were observed based on literary studies and based on this, the research hypothesis and research models were set. Then, through the operational definition of the variables were made and questionnaires were written and distributed, while the collected surveys were used to for positive analysis. Analysis methods used for the collected data was reliability analysis and factor analysis. In order to prove the research hypothesis, hierarchical regression analysis was used, and covariance structure analysis was used to prove the research model.
Samples for this study were extracted from employees in wholesale/retail and service companies in Seoul and Jeju Special Self-Governing Province. The period for the investigation was from November 1, 2007 to February 28, 2008. An interviewer distributed a total of 1,500 questionnaires through visitations. A total of 1,125 questionnaires were recollected and excluding non-diligent responses and questionnaires that have biased distribution, 971 were utilized for analysis.
Summary of the research results are as seen below.
First, three sub-dimensional behavior factors(confidence boost, humanly treatment, personal exchange behavior) were extracted for pygmalion leadership through observatory factor analysis and verification factor analysis, while three sub-dimensional behavior factors(duty behavior, innovation behavior and organization citizen behavior) were extracted for execution of duties.
Secondly, pygmalion leadership has a positive(+) effect on LMX quality, sense of self-efficacy and work motivation. In particular, the variation proportion that pygmalion leadership explains LMX quality was at 58.3%, thus showing that the it is a very important variable that must be considered as a preceding variable for the LMX quality. In addition, the three sub-factors(confidence boost, humanly treatment, personal exchange behavior) of pygmalion leadership and their distinguished relationship with LMX quality, sense of self-efficacy and work motivation were examined to confirm that in the LMX quality, there were three sub-behaviors with similar explanatory forces, thus showing that it had a relatively similar effect. However, for the sense of self-efficacy the leader's confidence boosting behavior and for work motivation, humanly treatment had higher effects compared to other leader behaviors.
Thirdly, results from the linear regression analysis showed that pygmalion leadership had a positive (+) effect on duty behavior, innovation behavior and organization citizen behavior. In addition, after verifying the distinguished relationship of sub-behaviors for pygmalion leadership and execution of duties, only confidence boosting behavior had an effect on duty behavior and for innovation behavior, confidence boosting behavior and social exchange behavior had a greater effect compared to humanly exchange behavior. Analysis also showed that in organization citizen behavior, confidence boosting behavior and humanly treatment behaviors had greater influence than personal exchange behaviors. However, when analyzing the structure model taking into consideration the effect relationship of all variables inserted into the research model, pygmalion leadership did not have a direct effect on the organization citizen behavior. It is presumed that the reason for this is because the sense of self-efficacy and work motivation that were selected as the variable media in the research model had high explanatory forces as a preceding variable for organization citizen behavior.
Fourth, linear regression analysis results showed that LMX quality had a positive(effect) on innovation behavior and organization citizen behavior and that a sense of self-efficacy had a positive (+) effect on variables for execution of duties(duty behavior, innovation behavior, organization citizen behavior), while work motivation had a positive effect on duty behavior and organization citizen behavior. However, results from the covariance analysis showed that LMX quality had a positive effect on duty behavior and innovation behavior, while a sense of self-efficacy and work motivation had a positive effect on the variables for execution of duties(duty behavior, innovation behavior, organization citizen behavior). Differences in linear regression analysis and covariance are that the effects of LMX quality on organization citizen behavior and the effects of work motivation on innovation behavior are different. Therefore, there was a need to come to a precise conclusion by combining the direct and indirect effects taking into consideration both interpretation of one-to-one relationships of the two variables, as well as several variables simultaneously.
Points to consider found based on the research results are as seen below.
First, the feasibility of the pygmalion leadership concept was examined to define the pygmalion behavior that could be used within the organization in order to provide an opportunity to expand pygmalion leadership research to the business administration and management sector. Pygmalion leadership studies have thus far been concentrated in the education sector. However, by presenting pygmalion leader behavior that could be utilized within the organization, this study provides an opportunity to perform additional studies using this in the management sector.
Second, the effects of pygmalion leadership was verified from multiple dimensions. Efficiency, satisfaction and flexible are needed to maintain and develop modern organizations. This study uses sense of self-efficacy and work motivation as variables for measuring satisfaction, duty behavior as variables for measuring efficiency, organization citizen behavior and innovation behavior as variables for measuring flexibility in the research model to confirm the multi-dimensional effects of pygmalion leadership.
Third, it was confirmed that pygmalion leadership was a variable that increased the explanatory force of LMX quality. LMX theories thus far did not explain the distinguished leader behavior directed towards constituents within a group. In other words, this does not present concrete and distinguished leader behaviors for improving LMX quality. This study presents leader behaviors that improve the explanatory force of LMX quality in order to supplement the areas that LMX quality research in the past overlooked. In particular, it was proven that pygmalion leader behaviors are variables that must be taken deeply into consideration for improvement of LMX quality.
Fourth, the distinguished effects of pygmalion leader behaviors on the sense of self-efficacy and work motivation were proven to present leader behaviors needed to increase the desire for executing duties and to enhance the judgment of value for future work execution capacities. Results showed that in order to increase the current desire for executing duties, behavior related to humanly treatment for subordinates was effected and in order to increase the value judgment on future work execution capacities, behaviors to boost the confidence of subordinates were more effective.
Fifth, by revealing the structural relationship in which pygmalion leadership effects the execution of duties as a partial medium on the subordinate's sense of self-efficacy and work motivation, I have presented a plan for monitoring the pygmalion leadership effects on site in the organization. In order to enhance the effects of pygmalion leadership on the subordinate's work performance, the subordinate's emotional response to work must be stirred and through this emotional response, the sense of self-efficacy and work motivation should be take into consideration. In addition, because these emotional responses are variables that can maintain the continued effect of behavior, it must be monitored to check whether pygmalion leadership is continuously applied to these factors.
Despite the above points to take into consideration, this study includes several limitations.
First, in this study the pygmalion leader behavior appears in three factor structure and in the studies of Oh In-su and Seo Yong-weon(2001) using the same survey questions, it appeared in a four factor structure. This means that there is the possibility that the factor structure can change depending on the subject of the research. Hence, there must be continuous studies made on developing measuring tools with high possibility for use in the management sector.
Second, the measurement variables composing the factors used in linear regression analysis and the measurement variables composing the factors used in the structure formula are slightly different. This is because results of the confirmation factor analysis showed that 1 question for sense of self-efficacy, 1 question for work motivation and 1 measurement variable of the organization citizen behavior acts as a variable that interferes with focus feasibility. Hence, there were fine differences between the research results of the linear regression analysis and that of covariance structure analysis.
Third, taking into account that the identical respondent presented all information for the causes and result variables, change methods have identical problems and thus, we cannot exclude the possibility that it could create more causes than the actual cause.
Fourth, it was unable to overcome its limitations as a transverse study. Accurate verification of the cause and effect relationship could be obtained only through time-series data or tests, and therefore, it will be necessary to supplement the results of this study through this kind of research.
A few suggestions for future studies on pygmalion leadership are as seen below.
First, there is a need to develop pygmalion leader behavior measurement tools that could be generally used on site at the organization. Because there are very few studies that extend pygmalion leadership into the business sector, there has not yet been a measuring tool developed that could be used in a wide range. Therefore, there must first be in-depth research on this as well.
Second, there is a need to pursue the combination of leadership theories other than that of LMX theories, including the various leader behaviors that could effect the LMX quality aside from that of the pygmalion leader behavior. By doing so, it will be able to verify various ALS(Average leadership style) theories that could come into effect in the LMX.
Third, through studies on the similarity and differences of leader behaviors activated in leadership other than pygmalion leader behavior, there is a need to work on integrating the various leadership theories with new genres. With the growing uncertainties of the management environment, many leadership theories are appearing, while leader behavior types that come into effect from these theories are being introduced. However, the leader behavior presented by these theories repeat considerably similar patterns. Therefore, there is a need to improve the leader behavior patterns by revealing the similarities and differences of the many leader behaviors. By doing so, it will be easier for leadership education and utilization in the actual work environment.

본 연구는 그동안 교육학과 심리학에서 연구되어 왔던 피그말리온 효과를 기업현장의 리더행동들과 연결하여 접근하고자 하였다. 첫째, 피그말리온 리더행동들과 다차원의 직무수행 변수 즉, 과업행동, 혁신행동, 조직시민행동의 집중타당성을 검증하는 것이다. 둘째, 피그말리온 리더행동들과 LMX 질 및 부하의 정서적 반응(자기효능감, 직무동기) 간의 선형관계와 차별적 관계를 검증하는 것이다. 셋째, 피그말리온 리더행동들과 직무수행(과업행동, 혁신행동, 조직시민행동)간의 선형관계와 차별적 관계를 검증하는 것이다. 넷째, LMX 질 및 부하의 정서적 반응(자기효능감, 직무동기)과 직무수행(과업행동, 혁신행동, 조직시민행동)간의 선형관계를 검증하는 것이다. 다섯째, 피그말리온 리더십, LMX 질, 자기효능감, 직무동기, 과업행동, 혁신행동, 조직시민행동 간의 구조적 관계를 밝히는 것이다.
연구결과를 요약하면 다음과 같다. 첫째, 피그말리온 리더십은 탐색적 요인분석 및 확인요인 분석을 통해 3개 차원의 하위행동요인(자신감 고취 행동, 인간적 대우 행동, 사적교류 행동)이 추출되었고, 직무수행은 3차원의 하위요인(과업행동, 혁신행동, 조직시민행동)이 추출되었다.
둘째, 피그말리온 리더십은 LMX 질, 자기효능감, 직무동기에 정(+)의 영향을 미치고 있으며, 특히 피그말리온 리더십이 LMX 질을 설명하는 분산비율이 58.3%로 나타나, LMX 질의 선행변수로 매우 중요하게 고려되어야 할 변수임이 확인되었다. 또한 피그말리온 리더십의 3개 하위요인(자신감 고취 행동, 인간적 대우 행동, 사적교류 행동)이 LMX 질, 자기효능감, 직무동기에 미치는 차별적 관계를 검증한 결과, LMX 질에는 3개 하위행동이 비슷한 설명력을 보여, 비교적 균등한 영향을 미치고 있는 것으로 확인되었다. 그러나 자기효능감에는 리더의 자신감 고취 행동이, 직무동기에는 인간적 대우 행동이 다른 리더 행동에 비해 영향을 많이 미치는 것으로 확인되었다.
셋째, 선형회귀분석 결과, 피그말리온 리더십은 과업행동, 혁신행동, 조직시민행동에 정(+)의 영향을 미치고 있다. 또한 피그말리온 리더십의 하위행동들과 직무수행의 차별적 관계를 검증한 결과, 과업행동에는 자신감 고취 행동만이 영향을 미치고, 혁신행동에는 자신감 고취 행동과 사적교류 행동이 인간적 대우 행동보다 큰 영향을 미치며, 조직시민행동에는 자신감 고취 행동과 인간적 대우 행동이 사적교류 행동보다 큰 영향을 미치는 것으로 분석되었다. 그러나 연구모형에 투입된 모든 변수들의 영향관계를 고려한 구조모델을 분석한 결과에서는 피그말리온 리더십은 조직시민행동에 직접적 영향을 미치지 않는 것으로 나타났다. 이러한 이유는 연구모형에서 매개변수로 선정한 자기효능감과 직무동기가 조직시민행동의 선행변수로서 설명력이 높기 때문인 것으로 판단된다.
넷째, 선형회귀분석 결과 LMX 질은 혁신행동과 조직시민행동에 정(+)의 영향을 미치고, 자기효능감은 직무수행변수들(과업행동, 혁신행동, 조직시민행동)에 정(+)의 영향을 미치며, 직무동기는 과업행동과 조직시민행동에 정(+)의 영향을 미치고 있다. 그리고 공분산 구조분석 결과에서는 LMX 질은 과업행동과 혁신행동에, 자기효능감과 직무동기는 직무수행 변수들(과업행동, 혁신행동, 조직시민행동)에 정(+)의 영향을 미치고 있다. 선형회귀분석과 공분산구조분석의 차이는 LMX 질이 조직시민행동에 미치는 영향과 직무동기가 혁신행동에 미치는 영향이 다르다는 점이다. 따라서 두 변수 간의 일대일 관계의 해석과 여러 변수를 동시에 고려한 직·간접 효과를 혼합하여 정교한 결론을 내릴 필요가 있다.
연구결과를 토대로 도출된 시사점을 제시하면 다음과 같다.
첫째, 피그말리온 리더십의 개념 타당성을 검증하여, 조직 내에 활용 가능한 피그말리온 행동들을 규명함으로써 피그말리온 리더십연구를 경영학 분야로 확장시키는 계기를 마련하였다. 그동안 피그말리온 리더십연구는 주로 교육학 분야에 집중되어 왔으나, 경영조직 현장에서 활용 가능한 피그말리온 리더행동을 제시함으로써 향후 경영학 분야에서 이를 활용한 추가적인 연구를 수행할 수 있는 계기를 마련하였다.
둘째, 피그말리온 리더십의 효과를 다차원으로 검증하였다. 현대조직의 유지·발전을 위하여 필요한 것은 효율성, 만족성, 유연성 확보이다. 본 연구는 만족성을 측정할 수 있는 변수로 자기효능감과 직무동기, 효율성 측정변수로 과업행동, 유연성 측정 변수로 조직시민행동과 혁신행동을 연구모형의 변수로 투입함으로써 피그말리온 리더십의 다차원 효과를 확인할 수 있었다.
셋째, 피그말리온 리더십이 LMX 질의 설명력을 높이는 변수임을 확인하였다. 그간의 LMX이론은 집단 내 구성원들을 향한 차별적 리더행동에 대한 설명이 없었다. 다시 말해 LMX 질 향상을 위한 구체적이고 차별적인 리더행동들을 제시하지 않았다. 본 연구는 LMX 질의 설명력을 높이는 리더행동들을 제시함으로써 그동안의 LMX 질 연구에서 간과된 부분을 보완하였다. 특히 피그말리온 리더행동들은 LMX 질 향상을 위해 매우 중요하게 고려되어야 할 변수임을 입증하였다.
넷째, 피그말리온 리더행동들이 자기효능감 및 직무동기에 미치는 차별적 효과를 입증하여, 현재 직무수행 욕구를 증대시키는데 필요한 리더행동과 미래 직무수행 능력에 대한 가치판단을 증대시키는데 필요한 리더행동들을 차별적으로 제시하였다. 연구결과 현재 직무수행 욕구를 증대시키기 위해서는 부하들에 대한 인간적 대우 행동과 관련된 행동들이 효과적이며, 미래 직무수행 능력에 대한 가치판단 증대를 위해서는 직무에 대한 부하들의 자신감을 고취시키는 행동이 더 효과적임을 입증하였다.
다섯째, 피그말리온 리더십이 부하의 자기효능감과 직무동기를 부분 매개로 직무수행에 미치는 구조적 관계를 밝힘으로써 경영조직 현장에서 피그말리온 리더십 효과를 모니터링하는 방안을 제시하였다. 피그말리온 리더십이 부하의 직무성과에 미치는 영향을 증대시키기 위해서 부하의 직무관련 정서적 반응을 촉발시켜야 하며, 이러한 정서적 반응으로 자기효능감, 직무동기 등을 고려할 필요가 있다. 또한 이러한 정서적 반응들은 행위의 지속효과를 유지할 수 있는 변수들이기 때문에 피그말리온 리더십이 이러한 변수들에 지속적으로 작용하는지를 모니터링하여야 한다.