Theory‐Testing, Generalization, and the Problem of External Validity

External validity refers to the generalization of research findings, either from a sample to a larger population or to settings and populations other than those studied. While definitions vary, discussions generally agree that experiments are lower in external validity than other methodological approaches. Further, external validity is widely treated as an issue to be addressed through methodological procedures. When testing theories, all measures are indirect indicators of theoretical constructs, and no methodological procedures taken alone can produce external validity. External validity can be assessed through determining (1) the extent to which empirical measures accurately reflect theoretical constructs, (2) whether the research setting conforms to the scope of the theory under test, (3) our confidence that findings will repeat under identical conditions, (4) whether findings support the theory being tested, and (5) the confirmatory status of the theory under test. In these ways, external validity is foremost a theoretical issue and can only be addressed by an examination of the interplay between theory and methods.     

You Thought That This Would Be Easy? Seeking an Understanding of Episodic Volunteering

VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations - This article is intended as the leading article in a Special Issue of Voluntas devoted to episodic volunteering from a...     

Exponentially weighted moving average control schemes with variable sampling intervals

The idea of using non-constant sampling intervals has been of interest in quality control applications since it was first suggested for the “skip-lot sampling plan” of Dodge. Recent interest has focused on the use of variable sampling interval (VSI) control schemes. VSI control schemes use a short sampling interval is given     

Bootstrap bias corrections for ensemble methods

This paper examines the use of a residual bootstrap for bias correction in machine learning regression methods. Accounting for bias is an important obstacle in recent efforts to develop statistical inference for machine learning. We demonstrate empirically that the proposed bootstrap bias correction can lead to substantial improvements in both bias and predictive accuracy. In the context of ensembles of trees, we show that this correction can be approximated at only double the cost of training the original ensemble. Our method is shown to improve test set accuracy over random forests by up to 70% on example problems from the UCI repository.     

Students’ Vocational Choices and Voluntary Action: A 12-Nation Study

Previous research on student involvement suggested that business and engineering students manifest lowest rates of voluntary action. Similarly, it was thought that social science students are the most involved in voluntary action, with students of natural sciences and humanities in the middle. However, there were very few studies that empirically compared these assertions. Furthermore, these assertions were not investigated from cross-cultural perspectives. Based on a study of students in 12 countries (N = 6,570), we found that even when controlling for background variables, social science students are actually less engaged in voluntary action than other students. Engineering students are higher than expected on voluntary action while students of humanities are the most involved in voluntary action. When studying these differences in the 12 selected countries, local cultures and norms form different sets of findings that suggest that there is no universal trend in choice of academic field and voluntary action.     

Intergenerational income immobility in Finland: contrasting roles for parental earnings and family income

An intergenerational model is developed, nesting heritable earning abilities and credit constraints limiting human capital investments in children. Estimates on a large, Finnish data panel indicate very low transmission from parental earnings, suggesting that the parameter of inherited earning ability is tiny. Family income, particularly during the phase of educating children, is shown to be much more important in shaping children’s lifetime earnings. This influence of parental incomes on children’s earnings rises as the children age because the returns to education rise. Despite Finland’s well-developed welfare state, persistence in economic status across generations is much higher than previously thought.     

A Nordhaus-Gaddum-type result for the induced path number

The induced path number ??(G) of a graph G is defined as the minimum number of subsets into which the vertex set of G can be partitioned so that each subset induces a graph. A Nordhaus-Gaddum-type result is a (tight) lower or upper bound on the sum (or product) of a parameter of a graph and its complement. If G is a subgraph of H, then the graph H?E(G) is the complement of G relative to H. In this paper, we consider Nordhaus-Gaddum-type results for the parameter ?? when the relative complement is taken with respect to the complete bipartite graph K _n,n .     

Competition network as a source of competitive advantage: The dynamic capability perspective and evidence from China

The search for competitive advantage is the defining inquiry of strategic management research. In this study, we draw on the dynamic capability lens to develop a counterintuitive view that positions competitors of a firm as an important source of competitive advantage. We argue that a firm's competitors form a competition network from which it can collect information about innovative ideas, product market, and related industries. Such information helps it calibrate market opportunities, update the resource base, and, eventually, strengthen its competitive advantage. This positive effect of competition network on competitive advantage will reasonably be contingent upon the proactive information search by the firm. The empirical results based on the survey data of 631 Chinese firms strongly support our theoretical model. This study identifies another distinctive source of competitive advantage than industry context or organizational resources as well as advancing our understanding of competition network.