Homework Help is Here – Start Your Trial Now!
Business
Operations Management
summarize two pages

summarize two pages

Practical Management Science
Practical Management Science
6th Edition
ISBN:9781337406659
Author:WINSTON, Wayne L.
Publisher:WINSTON, Wayne L.
Chapter2: Introduction To Spreadsheet Modeling
Section: Chapter Questions
Problem 20P: Julie James is opening a lemonade stand. She believes the fixed cost per week of running the stand...
See similar textbooks
icon
Related questions
Question
summarize two pages
of leadership perceptions to each trait individually, as noted earlier in the section on analytic strategy, we estimated a latent factor in a SEM analysis to extract the common variance from the set of traits to determine its relationship with leadership perceptions. When this factor was estimated using all seven of the trait dimensions, we found that six of the seven traits loaded positively and significantly on the latent factor, with moderate to strong standardized factor loadings ranging from 0.57 (for sensitivity) to 0.85 (for intelligence). However, the factor loading for the trait of tyranny was low magnitude, negative and not statistically significant, S. Trichas et al. / The Leadership Quarterly 28 (2017) 317-333 A = -0.08, p = 0.319, indicating that it shared at best only minimal common variance with the other leadership-related traits. Based upon this finding, we modified our analytic strategy, so that in analyses testing Hypothesis 4 and all other subsequent hy- potheses involving perceptions of trait dimensions, two sets of effects were estimated: (a) for a factor based on the six traits, and (b) for the Tyranny dimension only. The next paragraph presents more details of the revised trait factor model that was based on using only the relevant six trait dimensions as indicators. The revised trait factor model allowed the six trait measures of sensitivity, intelligence, potency, dynamism, dedication, and masculinity to load on a single common factor. An additional remaining covariance between the two traits of dynamism and po- tency was quite strong and was also not surprising given the nature of these two traits, therefore we also allowed the unique- nesses of these two variables to covary (r= 0.40, p < 0.001). In the full sample, the overall model fit was x² (227) = 39.958, df = 8, p <0.0001, RMSEA = 0.133, CFI= 0.936, SRMR = 0.049. Standardized factor loadings of the six trait indicators ranged from 0.58 (for masculinity) to 0.88 (for intelligence). As mentioned in our earlier discussion, this model would not be considered to be well-fitting for conventional purposes (although the CFI and SRMR indicate adequate fit). However, we were not trying to fit all underlying covariance relationships, rather were trying to isolate the commonality associated with leadership in order to study its relations with other variables, therefore this model was found to be acceptable. This preliminary development of a baseline factor model thus provided us with a basis for testing Hypothesis 4. To test Hypothesis 4, we performed a multi-group SEM analysis, splitting the data set by the happy versus nervous condition and estimating the model in the two groups simultaneously. The model was specified to estimate the covariance relationships of the overall leadership perceptions variable with (a) the 6-indicator latent trait factor and (b) the tyranny trait measure. Support for the hypothesis would consist of finding that these two relationships with overall leadership perceptions differed across the happy and nervous condition, specifically, that they were stronger in the happy condition than in the nervous condition. As pre- dicted, the covariance of overall leadership perceptions with the latent trait factor was higher in the happy condition (cov = 1.58, r = 0.84, p < 0.001), than in the nervous condition (cov = 0.79, r = 0.66, p < 0.001). However, the covariance of overall leader- ship perceptions with the separate trait of tyranny did not show the expected pattern. Specifically, the covariance was stronger and had a positive sign in the nervous condition (cov = 0.63, r = 0.29, p = 0.005), compared to the happy condition where it had a negative sign and was weaker (cov = -0.46, r=-0.18, p = 0.051). To determine whether the differences in the covariance values for the two groups were statistically significant, we performed a multi-group equivalence test. In the unconstrained condition, the covariances with overall leadership perceptions were allowed to freely differ in the happy versus nervous group (however, the factor loadings for the trait perceptions factor were constrained to be equal across groups in both models, which somewhat degraded fit but ensured that we were looking at comparable factors). The fit of this unconstrained model was (227)= 185.136, df = 46, p <0.001. It was compared to a constrained condition which forced the two sets of covariances to be equal, (227) = 207.463, df = 48, p < 0.001. After performing the necessary adjustment for use of the MLR estimator, we found the constrained model fit was significantly worse than that of the unconstrained model, adja x² (227) = 25.643, df = 2, p < 0.001. Thus, Hypothesis 4 was supported for the traits included in the trait factor, indicating that overall leadership perceptions were more strongly related to the pattern of perceptions on trait dimensions in the happy compared to the nervous condition. This result implies that leadership perceptions provide a stronger constraint when assessing traits in the happy than in the nervous condition. However, Hypothesis 4 was not supported for the test of the relationship with tyranny. Test of ILT endorsement effects Hypothesis 5 proposed that differences among subjects in their level of endorsement of the ILT dimensions would influence overall leadership perceptions and perceptions of leadership-related traits. If supported, this hypothesis would help explain why we would have stronger effects in the happy condition, as subjects are expected to make greater use of their ILTS in rating leadership in that condition. We hoped to show that ILT endorsements, which were rated by participants prior to viewing any stimulus materials and thus were independent of the manipulation, predict both overall leadership perceptions (H5a) and 328 -14° Leader Facial Emotion Expression (Happy vs. Nervous) ILT Endorsement 327 05ns 40 Overall Leadership 77 Leadership Trait Impression Dimension Factor 07ns Fig. 2. Mediational model of leader facial emotion expression effects on leadership trait attributions, as mediated through overall leadership impressions. S. Trichas et al / The Leadership Quarterly 28 (2017) 317-333 perceptions on perceived trait dimension scores (H5b). When results were estimated in the full sample (i.e, going across the happy and nervous conditions), the prediction of overall leadership perceptions from ILT endorsement was not statistically signif
Transcribed Image Text:of leadership perceptions to each trait individually, as noted earlier in the section on analytic strategy, we estimated a latent factor in a SEM analysis to extract the common variance from the set of traits to determine its relationship with leadership perceptions. When this factor was estimated using all seven of the trait dimensions, we found that six of the seven traits loaded positively and significantly on the latent factor, with moderate to strong standardized factor loadings ranging from 0.57 (for sensitivity) to 0.85 (for intelligence). However, the factor loading for the trait of tyranny was low magnitude, negative and not statistically significant, S. Trichas et al. / The Leadership Quarterly 28 (2017) 317-333 A = -0.08, p = 0.319, indicating that it shared at best only minimal common variance with the other leadership-related traits. Based upon this finding, we modified our analytic strategy, so that in analyses testing Hypothesis 4 and all other subsequent hy- potheses involving perceptions of trait dimensions, two sets of effects were estimated: (a) for a factor based on the six traits, and (b) for the Tyranny dimension only. The next paragraph presents more details of the revised trait factor model that was based on using only the relevant six trait dimensions as indicators. The revised trait factor model allowed the six trait measures of sensitivity, intelligence, potency, dynamism, dedication, and masculinity to load on a single common factor. An additional remaining covariance between the two traits of dynamism and po- tency was quite strong and was also not surprising given the nature of these two traits, therefore we also allowed the unique- nesses of these two variables to covary (r= 0.40, p < 0.001). In the full sample, the overall model fit was x² (227) = 39.958, df = 8, p <0.0001, RMSEA = 0.133, CFI= 0.936, SRMR = 0.049. Standardized factor loadings of the six trait indicators ranged from 0.58 (for masculinity) to 0.88 (for intelligence). As mentioned in our earlier discussion, this model would not be considered to be well-fitting for conventional purposes (although the CFI and SRMR indicate adequate fit). However, we were not trying to fit all underlying covariance relationships, rather were trying to isolate the commonality associated with leadership in order to study its relations with other variables, therefore this model was found to be acceptable. This preliminary development of a baseline factor model thus provided us with a basis for testing Hypothesis 4. To test Hypothesis 4, we performed a multi-group SEM analysis, splitting the data set by the happy versus nervous condition and estimating the model in the two groups simultaneously. The model was specified to estimate the covariance relationships of the overall leadership perceptions variable with (a) the 6-indicator latent trait factor and (b) the tyranny trait measure. Support for the hypothesis would consist of finding that these two relationships with overall leadership perceptions differed across the happy and nervous condition, specifically, that they were stronger in the happy condition than in the nervous condition. As pre- dicted, the covariance of overall leadership perceptions with the latent trait factor was higher in the happy condition (cov = 1.58, r = 0.84, p < 0.001), than in the nervous condition (cov = 0.79, r = 0.66, p < 0.001). However, the covariance of overall leader- ship perceptions with the separate trait of tyranny did not show the expected pattern. Specifically, the covariance was stronger and had a positive sign in the nervous condition (cov = 0.63, r = 0.29, p = 0.005), compared to the happy condition where it had a negative sign and was weaker (cov = -0.46, r=-0.18, p = 0.051). To determine whether the differences in the covariance values for the two groups were statistically significant, we performed a multi-group equivalence test. In the unconstrained condition, the covariances with overall leadership perceptions were allowed to freely differ in the happy versus nervous group (however, the factor loadings for the trait perceptions factor were constrained to be equal across groups in both models, which somewhat degraded fit but ensured that we were looking at comparable factors). The fit of this unconstrained model was (227)= 185.136, df = 46, p <0.001. It was compared to a constrained condition which forced the two sets of covariances to be equal, (227) = 207.463, df = 48, p < 0.001. After performing the necessary adjustment for use of the MLR estimator, we found the constrained model fit was significantly worse than that of the unconstrained model, adja x² (227) = 25.643, df = 2, p < 0.001. Thus, Hypothesis 4 was supported for the traits included in the trait factor, indicating that overall leadership perceptions were more strongly related to the pattern of perceptions on trait dimensions in the happy compared to the nervous condition. This result implies that leadership perceptions provide a stronger constraint when assessing traits in the happy than in the nervous condition. However, Hypothesis 4 was not supported for the test of the relationship with tyranny. Test of ILT endorsement effects Hypothesis 5 proposed that differences among subjects in their level of endorsement of the ILT dimensions would influence overall leadership perceptions and perceptions of leadership-related traits. If supported, this hypothesis would help explain why we would have stronger effects in the happy condition, as subjects are expected to make greater use of their ILTS in rating leadership in that condition. We hoped to show that ILT endorsements, which were rated by participants prior to viewing any stimulus materials and thus were independent of the manipulation, predict both overall leadership perceptions (H5a) and 328 -14° Leader Facial Emotion Expression (Happy vs. Nervous) ILT Endorsement 327 05ns 40 Overall Leadership 77 Leadership Trait Impression Dimension Factor 07ns Fig. 2. Mediational model of leader facial emotion expression effects on leadership trait attributions, as mediated through overall leadership impressions. S. Trichas et al / The Leadership Quarterly 28 (2017) 317-333 perceptions on perceived trait dimension scores (H5b). When results were estimated in the full sample (i.e, going across the happy and nervous conditions), the prediction of overall leadership perceptions from ILT endorsement was not statistically signif
Fig. 2. Mediational model of leader facial emotion expression effects on leadership trait attributions, as mediated through overall leadership impressions. 328 S. Trichas et al/ The Leadership Quarterly 28 (2017) 317-333 perceptions on perceived trait dimension scores (H5b). When results were estimated in the full sample (i.e., going across the happy and nervous conditions), the prediction of overall leadership perceptions from ILT endorsement was not statistically signif- icant, B = 0.13, p = 0.081, thus failing to support H5a. However, support was found for H5b with a significant prediction of the trait dimensions factor by endorsement of ILTS, B = 0.21, p = 0.004. The parallel test predicting only the tyranny trait from ILT endorsement was not significant, 3= 0.08, p = 0.210. The related Hypothesis 6, which proposed a happy versus nervous condi- tion difference in the relationship of ILT endorsements to (a) the overall leadership perceptions, (b) the trait dimensions factor, and (c) tyranny, also was not supported, adjA² = 1.491, df = 3, p = 0.316. Exploratory mediational model Finally, to provide an integrative framework within which to think about our results, we tested a theory-based mediational model (see Fig. 2) which proposed that the effects of the facial emotion expression of the leader on trait ratings of the leader was influenced by ILTS and mediated through overall leadership impression. Based on the logic that perceivers use categories to make sense of and simplify complex data, and also that categorization can be indexed by a global rating (Lord et al., 1984), we expected that rapid interpretation of a stimulus reflected in ratings in terms of overall leadership would be an appropriate index of leader categorization. Yet it is also possible that trait ratings drive interpretations of facial emotions and are antecedent to leadership perceptions, which is an alternative mediational logic. Hence we examined a mediational model on an exploratory basis. The mediational model was also estimated using SEM, and it used the trait dimensions factor as the dependent variable. As a preliminary step, an alternative model that did not include the mediator was estimated. In this preliminary model, the path coef- ficient relating the leader facial emotion expression manipulation directly to the trait dimension factor was statistically significant, B = 0.34, p<0.001, establishing again that the manipulation did indeed influence trait ratings. In addition, in this model ILT en- dorsement also significantly predicted the trait dimension factor, ß = 0.18, p = 0.008. Next the model was re-estimated, this time including overall leadership perceptions as a potential mediator of the effects of the manipulation and of ILT endorsement. As shown in Fig. 2, when the mediating variable was included in the model, the re- maining direct effect for the leader facial emotion was reduced substantially from its previous value of 0.34, and was no longer statistically significant, B = 0.05, p = 0.40. Instead, leader facial emotion expression related significantly to the mediating variable of overall leadership impression, B = 0.40, p < 0.001, which in turn related significantly to the trait dimension factor, ß = 0.77, p<0.001. A Sobel test showed that this indirect effect of leader facial emotion expression on the trait dimension factor was sta- tistically significant, aß = 0.31, p<0.001. However, ILT endorsement effects were not carried through the mediator, although there was a significant but low magnitude direct effect of ILT endorsement on the trait factor, ß = 0.11, p = 0.007. Details of this model are displayed in Fig. 2. Discussion The purpose of the current study was to add to our knowledge of several dynamic aspects of leadership perception that pertained to rapid use of facial expressions in the formation of leadership impressions. Specifically, we examined the role played by participants' ILT endorsement and the emotion displayed by stimulus individuals, and traced their effects through overall lead- ership perceptions to perception trait ratings. The emotions expressed by the focal individual affected overall leadership percep- tions, perceptions on trait ratings, the intercorrelations of perceptions on trait ratings, and the relation of a general factor based on perceptions of trait ratings to overall leadership perceptions. All of these results are consistent with the greater use of schemas related to leadership in the happy compared to the nervous condition. Our study highlights that a minimal manipulation (here, one different frame in a series of pictures or a final section of a short video) was sufficient to induce different leadership percep- tions in perceivers. We also found significant effects of ILT endorsement on trait perception ratings, but these effects did not vary across the emotion manipulation. ILT endorsement did not significantly predict overall leadership perceptions. Dynamics of the underlying process Effects of emotions and faces Because our data were cross-sectional rather than longitudinal, we acknowledge that comments made in this discussion re- garding underlying dynamics are tentative. Yet our results are largely consistent with other research and with our underlying the- ory. Similar to Trichas and Schyns (2012), who found that smiling or happy faces are considered more leader-like, we found that experimentally manipulated emotions affected leadership ratings. The simplest interpretation of this result is that emotions are cues that indicate leadership per se, and perhaps leadership-related traits as well, to raters. This is consistent with our finding that the effects of emotions were relatively independent of ILT endorsement effects. Our results relating to the leader displaying a nervous facial expression are also consistent with Bono and Judge's (2004) finding that leaders high in neuroticism were rated lower on all dimensions of transformational leadership. In addition, they are also in line with Felfe and Schyns (2010), who showed that follower-rated leader neuroticism was negatively related both to perceptions of transformational leadership and af- fective commitment to the supervisor. In combination, these results imply that leaders perceived as emotionally unstable, either due to their facial expressions or as a general judgment of their personality, are perceived as less leader-like. 329 S. Trichas et al/ The Leadership Quarterly 28 (2017) 317-333 Such results suggest that ILTS may include prototypical emotions and/or emotional displays as well as behaviors, a possibility that has not yet been considered by IIT researchers and should be explored further In doing so it is essential for researchers to
Transcribed Image Text:Fig. 2. Mediational model of leader facial emotion expression effects on leadership trait attributions, as mediated through overall leadership impressions. 328 S. Trichas et al/ The Leadership Quarterly 28 (2017) 317-333 perceptions on perceived trait dimension scores (H5b). When results were estimated in the full sample (i.e., going across the happy and nervous conditions), the prediction of overall leadership perceptions from ILT endorsement was not statistically signif- icant, B = 0.13, p = 0.081, thus failing to support H5a. However, support was found for H5b with a significant prediction of the trait dimensions factor by endorsement of ILTS, B = 0.21, p = 0.004. The parallel test predicting only the tyranny trait from ILT endorsement was not significant, 3= 0.08, p = 0.210. The related Hypothesis 6, which proposed a happy versus nervous condi- tion difference in the relationship of ILT endorsements to (a) the overall leadership perceptions, (b) the trait dimensions factor, and (c) tyranny, also was not supported, adjA² = 1.491, df = 3, p = 0.316. Exploratory mediational model Finally, to provide an integrative framework within which to think about our results, we tested a theory-based mediational model (see Fig. 2) which proposed that the effects of the facial emotion expression of the leader on trait ratings of the leader was influenced by ILTS and mediated through overall leadership impression. Based on the logic that perceivers use categories to make sense of and simplify complex data, and also that categorization can be indexed by a global rating (Lord et al., 1984), we expected that rapid interpretation of a stimulus reflected in ratings in terms of overall leadership would be an appropriate index of leader categorization. Yet it is also possible that trait ratings drive interpretations of facial emotions and are antecedent to leadership perceptions, which is an alternative mediational logic. Hence we examined a mediational model on an exploratory basis. The mediational model was also estimated using SEM, and it used the trait dimensions factor as the dependent variable. As a preliminary step, an alternative model that did not include the mediator was estimated. In this preliminary model, the path coef- ficient relating the leader facial emotion expression manipulation directly to the trait dimension factor was statistically significant, B = 0.34, p<0.001, establishing again that the manipulation did indeed influence trait ratings. In addition, in this model ILT en- dorsement also significantly predicted the trait dimension factor, ß = 0.18, p = 0.008. Next the model was re-estimated, this time including overall leadership perceptions as a potential mediator of the effects of the manipulation and of ILT endorsement. As shown in Fig. 2, when the mediating variable was included in the model, the re- maining direct effect for the leader facial emotion was reduced substantially from its previous value of 0.34, and was no longer statistically significant, B = 0.05, p = 0.40. Instead, leader facial emotion expression related significantly to the mediating variable of overall leadership impression, B = 0.40, p < 0.001, which in turn related significantly to the trait dimension factor, ß = 0.77, p<0.001. A Sobel test showed that this indirect effect of leader facial emotion expression on the trait dimension factor was sta- tistically significant, aß = 0.31, p<0.001. However, ILT endorsement effects were not carried through the mediator, although there was a significant but low magnitude direct effect of ILT endorsement on the trait factor, ß = 0.11, p = 0.007. Details of this model are displayed in Fig. 2. Discussion The purpose of the current study was to add to our knowledge of several dynamic aspects of leadership perception that pertained to rapid use of facial expressions in the formation of leadership impressions. Specifically, we examined the role played by participants' ILT endorsement and the emotion displayed by stimulus individuals, and traced their effects through overall lead- ership perceptions to perception trait ratings. The emotions expressed by the focal individual affected overall leadership percep- tions, perceptions on trait ratings, the intercorrelations of perceptions on trait ratings, and the relation of a general factor based on perceptions of trait ratings to overall leadership perceptions. All of these results are consistent with the greater use of schemas related to leadership in the happy compared to the nervous condition. Our study highlights that a minimal manipulation (here, one different frame in a series of pictures or a final section of a short video) was sufficient to induce different leadership percep- tions in perceivers. We also found significant effects of ILT endorsement on trait perception ratings, but these effects did not vary across the emotion manipulation. ILT endorsement did not significantly predict overall leadership perceptions. Dynamics of the underlying process Effects of emotions and faces Because our data were cross-sectional rather than longitudinal, we acknowledge that comments made in this discussion re- garding underlying dynamics are tentative. Yet our results are largely consistent with other research and with our underlying the- ory. Similar to Trichas and Schyns (2012), who found that smiling or happy faces are considered more leader-like, we found that experimentally manipulated emotions affected leadership ratings. The simplest interpretation of this result is that emotions are cues that indicate leadership per se, and perhaps leadership-related traits as well, to raters. This is consistent with our finding that the effects of emotions were relatively independent of ILT endorsement effects. Our results relating to the leader displaying a nervous facial expression are also consistent with Bono and Judge's (2004) finding that leaders high in neuroticism were rated lower on all dimensions of transformational leadership. In addition, they are also in line with Felfe and Schyns (2010), who showed that follower-rated leader neuroticism was negatively related both to perceptions of transformational leadership and af- fective commitment to the supervisor. In combination, these results imply that leaders perceived as emotionally unstable, either due to their facial expressions or as a general judgment of their personality, are perceived as less leader-like. 329 S. Trichas et al/ The Leadership Quarterly 28 (2017) 317-333 Such results suggest that ILTS may include prototypical emotions and/or emotional displays as well as behaviors, a possibility that has not yet been considered by IIT researchers and should be explored further In doing so it is essential for researchers to
Expert Solution
trending now

Trending now

This is a popular solution!

steps

Step by step

Solved in 2 steps

SEE SOLUTIONCheck out a sample Q&A here
Blurred answer
Similar questions
Recommended textbooks for you
Practical Management Science
Practical Management Science
Operations Management
ISBN:
9781337406659
Author:
WINSTON, Wayne L.
Publisher:
Cengage,
Operations Management
Operations Management
Operations Management
ISBN:
9781259667473
Author:
William J Stevenson
Publisher:
McGraw-Hill Education
Operations and Supply Chain Management (Mcgraw-hi…
Operations and Supply Chain Management (Mcgraw-hi…
Operations Management
ISBN:
9781259666100
Author:
F. Robert Jacobs, Richard B Chase
Publisher:
McGraw-Hill Education
Business in Action
Business in Action
Operations Management
ISBN:
9780135198100
Author:
BOVEE
Publisher:
PEARSON CO
Purchasing and Supply Chain Management
Purchasing and Supply Chain Management
Operations Management
ISBN:
9781285869681
Author:
Robert M. Monczka, Robert B. Handfield, Larry C. Giunipero, James L. Patterson
Publisher:
Cengage Learning
Production and Operations Analysis, Seventh Editi…
Production and Operations Analysis, Seventh Editi…
Operations Management
ISBN:
9781478623069
Author:
Steven Nahmias, Tava Lennon Olsen
Publisher:
Waveland Press, Inc.
SEE MORE TEXTBOOKS