Comparison of ridge regression and neural networks in modeling multicollinear data Advisors/Committee Members: Gerth, Richard J.

▼ Using a work-role attachment perspective, this study examined the impact of work attitudes (i.e., job involvement, organizational commitment, and career commitment) on desires to extend career employment through phased retirement. Data were collected via questionnaires that were emailed to non-faculty university employees, and hypotheses were tested with correlational analyses and hierarchical regression analyses. Consistent with predictions, job involvement, organizational commitment, and two dimensions of career commitment (career identity and career resilience) demonstrated significant, positive bivariate relationships with preferences to work beyond the planned retirement age in phased retirement. Further, work attitudes collectively contributed to variation in phased retirement preferences, even after controlling for age, finance, and health. Consistent with previous research, some support was found for the influence of work attitudes on decisions regarding the timing of traditional retirement (as measured by the planned retirement age). Taken together, the results lend some support for the assumption that phased retirement may be used as a retention tool for dedicated workers. Future research is necessary to test whether these preferences are manifested in actual participation in a phased retirement program. Advisors/Committee Members: Griffeth, Rodger.

▼ Research investigating the role of emotion in cross-cultural experiences has received little attention by organizational researchers, despite its potential link to expatriate success and adjustment in the international assignment. The present study expands on work by Gullekson and Vancouver (in press), which examined perceptions of emotional display for international sojourners. Specifically, their research revealed the “guest effect” in which international participants reported that one should display less emotion in the host culture, despite the cultural norms for greater display of emotion the host culture than in sojourners' home cultures. The present study aimed to determine whether there are differential norms for the display of emotion between expatriates and Americans working in the U.S. Status as a key mechanism for the guest effect was also examined. Specifically, both American and international sojourners watched a video vignette of a workplace interaction in which the target character was either an American or an expatriate, and displayed either little or intense anger and happiness. To replicate the guest effect, participants then rated the appropriateness or the display with the expectation that participants would rate it more appropriate for the expatriate to display less emotion than the American worker. Participants also rated the status of the target character in the organization to determine if lower status was conferred to the expatriate than to the American. Results did not replicate the guest effect with either the American or international sample, but a status differential was found, such that the expatriate was rated as lower status than the American worker. However, this result was found with the international sample, not the American sample, implying a potential misperception on the part of the sojourners, rather than an actual status differential exists in the U.S. The theoretical implications as well as the practical applications for organizations sending individuals abroad are discussed. Advisors/Committee Members: Vancouver, Jeffrey.

▼ Researchers investigating the psychological contract in employment relationships have primarily focused on employee and work outcomes that are the result of psychological contract breach and/or psychological contract violation. However, little empirical research has been done investigating the psychological contract formation process. The current study investigated the psychological contract of individuals in the department-participant relationship of a university experimental research program. Specifically, the study assumed that research participants do hold psychological contracts regarding the department-participant relationship. When forming these contracts,participants incorporated information from sources other than the department in their psychological contracts 32% of the time. Additionally, when asked to identify the source of these obligations, 30% of the time participants identified non-departmental items as having been presented by the department; thus, there is evidence that participants make source misattributions. Finally, implications for the work context and future directions for research are discussed. Advisors/Committee Members: Vancouver, Jeffrey B.

▼ The body of literature surrounding the study of motivation is often criticized for the plethora of research that has emerged from varying theoretical perspectives. The purpose of this dissertation was to test part of an integrative theory of the role of values in motivation based on perceptual control theory. Specifically, the values construct was reviewed using the process-content framework. The review concluded that values have been treated in two ways in motivation theory: a) as desired end-states (i.e., goals) and b) as a level of attractiveness or satisfaction. An empirical examination of the values as goals hypothesis based on perceptual control theory was conducted. It was expected that participants would learn to associate a task for which they received praise with the value of being well-respected (i.e., instrumentality). Additionally, receiving praise for performance on a task should influence the anticipated value (i.e., valence) of completing a second task for which participants previously learned they would receive praise and that receiving praise for performance on a task would influence participant's future task choice. Finally, the study explored whether receiving praise on a task prior to learning influenced the strength of a learned association. Though findings demonstrate support for a learned association between a task and value system (Hypothesis 1), I failed to find support for the remaining hypotheses. Thus, I found no evidence to support the proposition that values function as higher-level goals. Both methodological and theoretical considerations that may have contributed to this lack of findings are discussed. Advisors/Committee Members: Vancouver, Jeffrey.

▼ Applied psychologists have long recognized the importance of measurement as a key component of research quality, but the use of psychometrically sound measurement practices has not kept pace. Recent evidence has emerged to suggest that weak measurement practices can have serious implications for the accuracy of parametric statistics. Two simulation studies (Embretson, 1996; Kang & Waller, 2005) have identified that response score scaling and assessment appropriateness heavily influence the Type I error rate for interaction effects in moderated statistical models when simple raw scores are used to operationalize a latent construct. However, the use of item response theory (IRT) models to rescale the raw data into estimated theta scores was found to mitigate these effects. The purpose of this dissertation was to generalize these results to polytomous data that is commonly found in applied psychological research using a Monte Carlo simulation. Consistent with the previous studies, inflated Type I error rates for the interaction effect in a moderated multiple regression model were observed when raw scores were used to operationalize a latent construct. In the most extreme cases, this inflation approached 85%. Also consistent with previous studies, psychometric factors were found to have a greater impact on raw scores than on estimated theta scores, and assessment appropriateness was found to be the most influential factor on the empirical Type I error rate. Inconsistent with previous studies, an inflated Type I error rate was also observed under some conditions for the estimated theta scores suggesting that the graded response model (GRM) may not have provided a sufficiently equal-interval metric. Additionally, the expected interaction between assessment appropriateness and assessment fidelity was not found to be significant. Overall, these results suggest that the IRT-derived scores were more robust to spurious interactions than simple raw scores, but may still result in inflated Type I error rates under some conditions. The implications of these results are discussed from two perspectives. The performance of the GRM under the simulated conditions is emphasized for measurement researchers, and the usefulness of model-based measurement practices for improving research quality is emphasized for applied psychologists. Advisors/Committee Members: Griffeth, Rodger W.

▼ Sexual harassment (SH) in organizations is a complex problem that has been of interest for nearly 30 years. A number of theories have been proposed to explain SH, and while none of them have had extensive support, power is a consistent element. Both traditional and non-traditional forms of SH exist, including contrapower harassment in which the harasser appears to hold less formal power than the victim. The purpose of this study was to investigate perceptions of contrapower SH in order to gain a more thorough understanding of SH. Participants read one of 8 vignettes, in which several characteristics had been varied. Participants answered questions regarding whether harassment had occurred, and if so, who in the interaction was at fault. Results revealed evidence for both traditional and contrapower SH. However, no significant findings for participant gender or exposure to training/education were found. Advisors/Committee Members: Popovich, Paula M.

▼ The purpose of this research was to assess the existence of attitudes and connotations toward email names, as well as to apply a theoretical framework for describing the process of how those attitudes may influence job related evaluations. Results from 2 studies using undergraduates confirmed the existence of various attitudes toward different email name-types, as well as the influence of those perceptions on evaluations of various selection criteria. These results suggested that individuals need to be aware of the potential impressions others have of them based on their email names. However, while results indicated that email name-type did not have an influence on interview decisions, the quality of individuals’ resumes did affect whether or not an applicant was asked to return for an interview. Advisors/Committee Members: Popovich, Paula M.

▼ Sabotage is a severe behavior that has professional and legal ramifications for both the saboteur and the organization. Currently, research investigating sabotage has relied on self-report methodology to measure retaliation behavior. However, self-report of negative behavior can result in severe consequences for employees (i.e., saboteurs), raising the question of respondent honesty. To address this issue, an objective measure of employee reactions was developed and implemented in a work simulation task to better understand employee reactions. This study found that participants who experienced a single occurrence of injustice would engage in sabotage behaviors. This finding supports the current literature and provides new directions and methods for assessing sabotage and other counterproductive work behaviors. Advisors/Committee Members: Popovich, Paula.

▼ The zeitgeist of psychology and economics has been that financial incentives lead to increases in effort. When examined empirically under static conditions, research shows that increases in value do lead to increases in motivation. What is not clear is how financial incentives influence motivation dynamically. A number of field studies have investigated individuals work motivation over time and found that individuals work more when their wage rate is lower, not higher. The authors of these studies theorize that individuals are using daily income goals and stop working once they reach their goal. However, these studies have largely been cross-sectional. This study seeks to test experimentally the direct influence of goals on motivation in a dynamic performance context. Results from the current study indicate that the majority of individuals worked longer when their wage rate was high and less when the wage rate was low, even when individuals had a monetary goal for each trial. However, individuals in the monetary goal condition who were loss averse and reported not valuing money highly did work less when their wage rate was higher. Advisors/Committee Members: Vancouver, Jeffrey.