Hierarchical linear modeling as an example for measuring change over time in a leadership development evaluation context

One of the difficulties in evaluating leadership development is measuring whether and how people change over the time period of the leadership development initiative. Even when change over time is an inherent part of the design and evaluation of leadership development, events may occur outside of the control of evaluators that limit the effectiveness of adequately and accurately assessing change over time. With data from a leadership development initiative designed to account for change over time, this article suggests hierarchical linear modeling (HLM) as a multilevel methodological technique to assess change over time in a leadership development context. This article will use real change over time data from a leadership development initiative and discuss the logic and rationale of HLM. We use HLM as an example of a multilevel methodological tool to investigate typical change over time questions in leadership development evaluation.     

Quantitative and qualitative examination of propositions concerning supervisor–subordinate convergence in descriptions of leader–member exchange (LMX) quality

Two studies were conducted to quantitatively and qualitatively examine supervisor–subordinate differences in perceptions concerning the quality of their LMX relationship. The results provided strong support for a lack of construct convergence between these two perspectives. Specifically, the findings suggest that leaders tend to focus more on task-oriented subdimensions and tend to give them greater weight when evaluating LMX relationships. However, subordinates are more oriented toward the social aspects of the relationship and perceive them as more important. This differential pattern was found to be generalizable across gender, race, and industry types.     

Simple methods for evaluating and comparing binary experiments

We consider a confidence parametrization of binary information sources in terms of appropriate likelihood ratios. This parametrization is used for Bayesian belief updates and for the equivalent comparison of binary experiments. In contrast to the standard parametrization of a binary information source in terms of its specificity and its sensitivity, one of the two confidence parameters is sufficient for a Bayesian belief update conditional on a signal realization. We introduce a confidence-augmented receiver operating characteristic for comparisons of binary experiments for a class of “balanced” decision problems, relative to which the confidence order offers a higher resolution than Blackwell’s informativeness order.

Where observation is concerned, Chance favors only the prepared mind. —Louis Pasteur (1822–1895).

     

Efficiency determinants in retail stores: a Bayesian framework

This paper analyzes the impact of vertical integration, age, geographic expansion, and low price strategy on the cost efficiency of retail stores. We test our hypotheses using the innovative Bayesian frontier methodology. The data involve a sample of Spanish retail stores that operate in a highly competitive and dynamic environment. From the results, it is clear that cost efficiency is higher for stores that have: longer years in business, stronger geographical presence, and lower price offerings. Vertical integration, on the other hand, is negatively related to efficiency. Further discussions of these findings and related managerial implications are provided in the paper.     

Skill and education effects on earnings in 18 Countries: The role of national educational institutions

This study investigates whether the mechanisms why education is rewarded vary across countries. Do educational institutions affect the likelihood that support for a particular mechanism is found? Combining IALS survey data and OECD statistics on educational institutions, it was shown that the effect of measured skill on earnings - controlled for educational attainment - is lower in countries where educational institutions produce skills relevant for work through the vocational system. This indicates that the human capital perspective on education works particularly well in vocationally oriented educational systems, as the skills generated in education are strongly overlapping with the skills that are rewarded. An alternative mechanism sees education as a means for social closure through credentialization. Under the credentialization model, education is not primarily rewarded for the productivity-enhancing skills it entails, but rather for reasons unrelated to productivity. Following this theory education is used for selection into the organization, after which directly observable skills are determining wages. Assuming that a strongly differentiated educational system creates boundaries between social groups, it is hypothesized that strongly differentiated systems lead to stronger measured skill effects. We do not find support for this hypothesis.     

The value of a name: Estimating the economic impact of public land designation

Public land designations have been shown to have significant impact on their contextual regional economies. This paper reviews the existing literature on how public lands impact the proximate regional economies in which they are situated. Based on this collected wisdom the paper synthesizes past lessons into a comprehensive impact methodology. This methodology distinguishes quality effects in terms of visitors’ willingness to pay for sites of higher quality and quantity effects in terms of additional visitation to such sites to give a more comprehensive methodology for characterizing visitation and economic impact of protected area designation. We focus on National Monuments and National Parks, given that these two types of public lands have the broadest popular exposure and feature the most reliable data. A case study is then presented for a hypothetical protected area in Ferry County, Washington, USA, applying this new methodology to assess the likely economic effects of different types of protected areas.     

MULTIVARIATE DECOMPOSITION FOR HAZARD RATE MODELS

We develop a regression decomposition technique for hazard rate models, where the difference in observed rates is decomposed into components attributable to group differences in characteristics and group differences in effects. The baseline hazard is specified using a piecewise constant exponential model, which leads to convenient estimation based on a Poisson regression model fit to person-period, or split-episode data. This specification allows for a flexible representation of the baseline hazard and provides a straightforward way to introduce time-varying covariates and time-varying effects. We provide computational details underlying the method and apply the technique to the decomposition of the black-white difference in first premarital birth rates into components reflecting characteristics and effect contributions of several predictors, as well as the effect contribution attributable to race differences in the baseline hazard .