RELATIONSHIP BETWEEN NURSES’ LEADERSHIP STYLES AND POWER BASES

O objetivo deste estudo foi comprovar empiricamente a relação entre as bases de poder do líder e os estilos de liderança das enfermeiras. Trata-se de estudo quantitativo com amostra aleatória de 204 participantes, profissionais de enfermagem de um hospital público. Os instrumentos de mensuração foram: o SBDQ (Supervisory Behavior Description Questionaire) para identificar os estilos de liderança e o Perfil de Percepção do Poder para determinar os tipos de poder utilizados pelos líderes. Foi realizada análise descritiva, bivariada e multivariada. Com base nos resultados alcançados, verificou-se a relação proposta pela TLS (Teoria da Liderança Situacional) entre o poder coercitivo e o estilo de liderança E1 (determinar) e entre o poder referente e o estilo de liderança E3 (participar). Em outros casos, os resultados têm sido opostos aos esperados: a utilização do poder proposto pelo modelo diminui a probabilidade para desempenhar o estilo de liderança prescrito.

This quantitative study aimed to empirically evidence the relationship between the power bases of the leader and the leadership styles of nurses.The random sample consisted of 204 nursing professionals from a public hospital.The following measurement instruments were used: the SBDQ (Supervisory Behavior Description Questionnaire) to identify leadership styles and the Power Perception Profile to determine the types of power used by leaders.Descriptive, bivariate and multivariate analyses were used.Based on the results, two relationships proposed by the SLT (Situational Leadership Theory) were verified: between coercive power and S1 leadership style (telling), and between referent power and S3 leadership style (participating).In other cases, results have been opposite to expectations: the use of power proposed by the model decreases the probability of performing the prescribed leadership style.DESCRIPTORS: nursing staff; power; leadership; hospital administration

Com base nos resultados alcançados, verificou-se a relação proposta pela TLS (
This theory defines leadership as "the process of influencing the activities of an individual or a group in efforts towards goal achievement in a given situation".As its name states, the situation is particularly relevant in this theory.Previously, in other theoretical approaches, the effective leader was considered to have a set of specific characteristics or attributes.In the same way, other approaches focused on identifying the appropriate behavior and style in any situation.
The SLT, without invalidating the previous one, emphasizes the adjustment between leaders' behavior and situational demands.
The SLT recognizes that each leader can have a preferred style, however, the effectiveness of leadership depends on the extent to which the leader is able to adapt or modify his/her style when the situation requires.
Authors of the SLT (1) identify the leadership style based on two dimensions of the leader's conduct: they define the task behavior as all conducts forwarded to detail the task and functions of each of its members.In 1967, the 3D Leadership Theory was presented (4) , introducing effectiveness as the third dimension.Under this viewpoint, relationship and task conducts would not be sufficient, and effectiveness would depend on the degree of adjustment between the leader's style and the circumstance of the situation.
Authors of the SLT (5) previously adopted the idea of a third dimension to achieve leadership effectiveness.Of all possible variables that configure the situation, they identified the subordinates' maturity (or preparation) as the most relevant situational variable.
According to this model, there is no optimal leadership style to influence employees more According to the authors of the SLT (1) , each level of readiness corresponds to a certain leadership style.Thus, level 1 of readiness corresponds to S1, level 2 corresponds to S2 and so on.
The great appeal and dissemination of the model among leaders is well known (6) .According to several researchers (7) , the probability of successful leadership increases when the style reflects the appropriate power base.
This study aimed to verify whether the power bases used correspond to the theory, which led to the need of defining and describing what power is.
Power has been defined (8)(9) as the potential an individual has to influence another.Leadership is any effort exerted to influence and power is its potential influence, the resource that permits influencing.
On the other hand, some researchers (7) support the idea that it is not the leader's power that permits influencing his/her followers, but actually the perception they have of the leader's power.
Different classification systems of power bases have been proposed.Among them, the classification below is the most disseminated.It (10) identifies five bases of power.
Coercive power -is the perception of the subordinate on the leader's capacity to enforce punishments.
Reward power -subordinates recognize the leader's capacity to offer gratifications.
Legitimate power -this power is related to the leader's position or function.
Referent power -the leader inspires positive admiration and affection in subordinates.
Expert power -subordinates recognize the leader as someone with experience and ability.
A sixth power base was added later, the information power, which is the leader's ability to obtain relevant information for subordinates.
Years later, the set of power was configured with the addition of another type of power, the connection power (11) , which is defined as the subordinates' perception of the leader's ability to connect with influential people or organizations.
The SLT was finally completed by adding the relationship between the power bases and the leadership styles (7) .In this model, a specific match is prescribed between each of the power bases and the most appropriate style leaders should apply to exert the strongest possible influence on their followers.
These assumptions of the model are tested in this study.
In the last decades, nurses have occupied managerial jobs in the health system.This and other studies (12) provide useful scientific knowledge to the exercise of new positions and competences.

OBJECTIVE
This study aimed to empirically test the relationship between the different sources of power of the leader and nurses' management styles.
Thus, the following premises are presented: -the application of the coercive power base increases the probability of having S1 style (telling);

Measurement instruments
The SBDQ (Supervisory Behavior Description Questionnaire) was used to measure the leadership style (13) , as it provides data in two behavior dimensions (Initiating Structure and consideration, assimilable to the task and relationship behaviors, respectively).This instrument comes from studies on leadership at Ohio State University, United States.In recent decades, it has been extensively applied in research in different organizations.
SBDQ is a 48-item questionnaire that describes the behavior of the leader, using a 5-point scale in the answers.Among the items, 28 correspond to the dimension of Consideration and the others to the Initiating Structure dimension.As there is no SBDQ normative data parameter, the median was used to establish the cut-off point between the high and low values in each dimension.
The Power Perception Profile was used to measure the perception of power (14) .Each of the seven power bases is compared to the others, so as to obtain 21 pairs of compared statements.In each of the pairs, the interviewee has to assign three points aiming to obtain the score for each source of power.This instrument is distributed and sold in Spain by the Leadership Studies Center, Spanish Consultants, SA, adapted to Spanish from the English original.

Data analysis
First, descriptive analysis was carried out, with mean, median, standard deviation or percentages and frequencies, according to quantitative or qualitative variables, respectively.
Later, bivariate analysis through contrast of means using the ANOVA test was carried out.Finally, ten logistic regressions were estimated to verify the probability of using a determined leadership style, according to the different types of power.

Descriptive analysis
Table 1 shows the mean scores interviewees attributed to the power base used by leaders.Scores can range from 0 to 18. Results were subsequently analyzed in relation to the leadership styles through the following procedure.
Firstly, the scores of the 28 first items of the SBDQ were summed, to determine the score in the relationship dimension.
Secondly, the results of the last 20 items were summed to obtain the score in the task dimension.
Thirdly, the median of the relationship dimension, 65, and of the task dimension, 36, were used to classify the leaders with high and low scores in the dimensions.
The score 65 in the relationship dimension was included in the high category.The score 36, obtained in the task dimension, was included in the high category.
The median divided leaders into two categories in both dimensions: -low in the relationship dimension, identified as number 1; -high in the relationship dimension, identified as number 2; -low in the task dimension, identified as number 1; -high in the task dimension, identified as number 2.
Then the different categories were recoded according to the four leadership styles defined by the model.
Table 2 shows the distribution of the different styles.As observed in Table 3, the differences of the mean scores obtained in the different leadership styles are statistically significant in the coercive, connection, referent, information and expert types.

Multivariate analysis
Aiming to verify the probability that, using the power base described by the model, the corresponding leadership style could be determined, ten logistic regressions were estimated, one for each power base and the corresponding leadership style defined by the model.Table 4 presents the results, according to each power base and corresponding leadership style.Table 4 -Logistic regressions for the types of power and the leadership style It can be observed in Table 4 that the results of the regression analysis of the following variables were not statistically significant: connection power and S1 style, reward power in relation to S2 style, legitimate power with S2 style and with S3 style E3 and, at last, information power with S3 style.
On the other hand, the results of the regression analysis of the coercive power with S1 leadership style are statistically significant, that is, for each additional point in leader's coercive power, the probability of having S1 leadership style increases 1.137 times in comparison to other styles.
effectively; the leadership style should depend directly on the preparation of the subordinates, on whom the leader intends to exert influence so as to achieve his/ her aims.The readiness of the subordinates is concretized in their ability and willingness.The first refers to the experience and skill of the individual to perform a given task, and the second to the motivation to perform it.This theoretical model considers that there are four leadership styles, resulting from the combination of task behavior and relationship behavior.These two dimensions are independent; four leadership styles emerge from them, described as follows.Style 1 (S1) -telling -is characterized by aboveaverage task behavior and below-average relationship behavior.Style 2 (S2) -selling -is characterized by both task and relationship behavior above average.Style 3 (S3) -participating -is characterized by above-average relationship behavior and belowaverage task behavior.Style 4 (S4) -delegating -both relationship behavior and task behavior below average.

-
the use of the connection power increases the probability of developing the S1 style (telling); -the use of the connection power increases the probability of using the S2 style (selling); -the application of the reward power increases the probability of having the S2 style (selling); -the use of the legitimate power increases the probability of developing the S2 style (selling); -the application of the legitimate power increases the probability of using the S3 style (participating); -the use of the referent power increases the probability of having the S3 style (participating); -the use of the information power increases the probability of using the S3 style (participating); -the application of the information power increases the probability of having the S4 style (delegating); -the use of the expert power increases the probability of developing the S4 style (delegating).METHOD Type of study: descriptive.Participants Nursing professionals from a public hospital in Granada, Spain.The study was authorized by the Research Committee of the Virgen de las Nieves Hospital, responsible for complying with the ethical aspects of the research.Sample Simple random, with 290 out of 980 nursing professionals.The nursing professionals who were part of the sample were first located and then visited at their workplace.They were given a letter of presentation with the objectives of the research and asked to sign the free and informed consent to participate in this research.Those who accepted were informed how to collaborate and their anonymity was guaranteed.The percentage of answers was 77.6% (n=225), however, only 204 participants fully filled out the questionnaires.Of those, 53.9% were nurses, 7.9% were specialist technicians and the remaining 38.2% were nursing auxiliaries.In this sample, 11% of the participants was male.The average age was 44.37 years and standard deviation .62 years.The age interval ranged from 22 to 62 years.

Table 1 -
Description of the types of power

Table 3 -
Comparison of the average of the seven types of power in each of the leadership styles