Search results

Longitudinal regression analysis is conducted to clarify causal relations and control for unwanted influences from actor heterogeneity and state dependence on theoretically important coefficient estimates. Because strategic management contains theory on how firms differ and how firm actions are influenced by their current strategic position and recent experiences, consistency of theory and methodology often requires use of longitudinal methods. We describe the theoretical motivation for longitudinal methods and outline some common methods. Based on a survey of recent articles in strategic management, we argue that longitudinal methods are now used more frequently than before, but the use is still inconsistent and insufficiently justified by theoretical or empirical considerations. In particular, strategic management researchers should use dynamic models more often, and should test for the presence of actor effects, autocorrelation, and heteroscedasticity before applying corrections.

This article reviews the extensive history of dynamic performance research, with the goal of providing a clear picture of where the field has been, where it is now, and where it needs to go. Past research has established that job performance does indeed change, but the implications of this dynamism and the predictability of performance trends remain unresolved. Theories are available to help explain dynamic performance, and although far from providing an unambiguous understanding of the phenomenon, they offer direction for future theoretical development. Dynamic performance research does suffer from a number of methodological difficulties, but new techniques have emerged that present even more opportunities to advance knowledge in this area. From this review, I propose research questions to bridge the theoretical and methodological gaps of this area. Answering these questions can advance both research involving job performance prediction and our understanding of the effects of human resource interventions.

In this chapter, the authors consider the role of time for research in occupational stress and well-being. First, temporal issues in studying occupational health longitudinally, focusing in particular on the role of time lags and their implications for observed results (e.g., effect detectability), analyses (e.g., handling unequal durations between measurement occasions), and interpretation (e.g., result generalizability, theoretical revision) were discussed. Then, time-based assumptions when modeling lagged effects in occupational health research, providing a focused review of how research has handled (or ignored) these assumptions in the past, and the relative benefits and drawbacks of these approaches were discussed. Finally, recommendations for readers, an accessible tutorial (including example data and code), and discussion of a new structural equation modeling technique, continuous time structural equation modeling, that can “handle” time in longitudinal studies of occupational health were provided.

An individual’s driving style significantly affects overall traffic safety. However, driving style is difficult to identify due to temporal and spatial differences and scene heterogeneity of driving behavior data. As such, the study of real-time driving-style identification methods is of great significance for formulating personalized driving strategies, improving traffic safety and reducing fuel consumption. This study aims to establish a driving style recognition framework based on longitudinal driving operation conditions (DOCs) using a machine learning model and natural driving data collected by a vehicle equipped with an advanced driving assistance system (ADAS).

Specifically, a driving style recognition framework based on longitudinal DOCs was established. To train the model, a real-world driving experiment was conducted. First, the driving styles of 44 drivers were preliminarily identified through natural driving data and video data; drivers were categorized through a subjective evaluation as conservative, moderate or aggressive. Then, based on the ADAS driving data, a criterion for extracting longitudinal DOCs was developed. Third, taking the ADAS data from 47 Kms of the two test expressways as the research object, six DOCs were calibrated and the characteristic data sets of the different DOCs were extracted and constructed. Finally, four machine learning classification (MLC) models were used to classify and predict driving style based on the natural driving data.

The results showed that six longitudinal DOCs were calibrated according to the proposed calibration criterion. Cautious drivers undertook the largest proportion of the free cruise condition (FCC), while aggressive drivers primarily undertook the FCC, following steady condition and relative approximation condition. Compared with cautious and moderate drivers, aggressive drivers adopted a smaller time headway (THW) and distance headway (DHW). THW, time-to-collision (TTC) and DHW showed highly significant differences in driving style identification, while longitudinal acceleration (LA) showed no significant difference in driving style identification. Speed and TTC showed no significant difference between moderate and aggressive drivers. In consideration of the cross-validation results and model prediction results, the overall hierarchical prediction performance ranking of the four studied machine learning models under the current sample data set was extreme gradient boosting > multi-layer perceptron > logistic regression > support vector machine.

The contribution of this research is to propose a criterion and solution for using longitudinal driving behavior data to label longitudinal DOCs and rapidly identify driving styles based on those DOCs and MLC models. This study provides a reference for real-time online driving style identification in vehicles equipped with onboard data acquisition equipment, such as ADAS.

The purpose of this paper is to provide evidence that the U‐shaped relationship between intellectual property rights (IPRs) and per capita gross domestic product (GDP) observed in the past literature using a panel of data is not a consequence of longitudinal forces, as has been previously postulated, but instead a consequence of cross‐sectional influences.

Differences in the longitudinal and cross‐sectional relationship between IPRs and per capita GDP are analyzed through a variety of methods, including pooled regression analysis that isolates the regional differences that are critical in making an accurate longitudinal analysis from the panel data.

Analyzing the country data reveals that a longitudinal U‐shaped relationship is counterfactual, as countries generally do not weaken their IPRs once they are in place, barring a regime change or other alteration in their political economy. The significant U‐shape link between IPRs and per capita GDP empirically observed in preliminary analysis of the panel data is instead a result of cross‐sectional influences.

Making the distinction between the cross‐sectional and longitudinal relationship between IPRs and per capita GDP provides a more accurate insight about how IPRs change in a country as it develops.

The purpose of this paper is to explore the possibilities and pitfalls of longitudinal studies in the field of operations management (OM).

A longitudinal study conducted in Finland was analyzed from the methodological point of view. In the theoretical part, a framework for analyses was created by studying the literature of longitudinal studies within and without the OM field.

Longitudinal studies are important in revealing root‐cause effects. They are particularly important in generating new ideas and theories and in questioning old ones. However, longitudinal studies are somewhat difficult to conduct and some unexpected challenges may arise, such as the effect of technology development on data retrieval.

The analyses in this study are based on the data obtained in a longitudinal study conducted in the years 1993, 2004 and 2010. The sample in the longitudinal study is quite small for real quantitative statistical analysis, and therefore the conclusions made here are only indicative. However, the purpose of this study is to give some hints and guidance, and thus the lessons learned are valuable.

There are only a few studies describing methodological issues in longitudinal studies in the field of OM, and most of these studies are purely theoretical and do not refer to real experiences. This appears to be one of the first studies describing real experiences from a longitudinal study.

The purpose of this paper is to illustrate the value of extended time span coverage of state longitudinal education and workforce data system to inform and improve the effectiveness of future high impact expenditure decisions.

It used an analytical 29-year data file created by the author that links seven already-in-place education and workforce administrative record sources. Relying on the path dependency theory, multi-level mixed-effect logistic and multi-level mixed-effect linear regression models are used to test three hypotheses.

The findings are consistent with the hypotheses: inclusion of the multiple steps along a post-secondary education pathway and prior job histories are both critical to understanding workforce outcomes mechanisms; it takes time for the employment outcome effect to be evident and strong following education attainment.

The study concludes with research limitations and implications for decision makers to call for retaining and investing in administrative records with extended time span coverage, particularly for the already-in-place historical administrative records.

The paper is one of the first to demonstrate the value of extended time span coverage in a longitudinal state integrated data system through econometric modeling, using longitudinally integrated data linking seven administrative records covering continuously for 29 years. No matter for prior education or employment pathway, it is only through extended time span coverage that employment outcomes can be well measured and the rich nuances interpreting the mechanisms of education return on investment can be revealed.

Longitudinal investigations are often suggested but rarely used in operations and supply chain management (OSCM), mainly due to the difficulty of obtaining data. There is a silver lining in the form of existing large-scale and planned repeated cross-sectional (RCS) data sets, an approach commonly used in sociology and political sciences. This study aims to review all relevant RCS surveys with a focus on OSCM, as well as data and methods to motivate longitudinal research and to study trends at the plant, industry and geographic levels.

A comparison of RCS, panel and hybrid surveys is presented. Existing RCS data sets in the OSCM discipline and their features are discussed. In total, 30 years of Global Manufacturing Research Group data are used to explore the applicability of analytical methods at the plant and aggregate level and in the form of multilevel modeling.

RCS analysis is a viable alternative to overcome the confines associated with panel data. The structure of the existing data sets restricts quantitative analysis due to survey and sampling issues. Opportunities surrounding RCS analysis are illustrated, and survey design recommendations are provided.

The longitudinal aspect of RCS surveys can answer new and untested research questions through repeated random sampling in focused topic areas. Planned RCS surveys can benefit from the provided recommendations.

RCS research designs are generally overlooked in OSCM. This study provides an analysis of RCS data sets and future survey recommendations.

The purpose of this paper is to describe the longitudinal aerodynamic characterization of an unmanned cropped delta configuration from real flight data. In order to perform this task an unmanned configuration with cropped delta planform and rectangular cross-section has been designed, fabricated, instrumented and flight tested at flight laboratory in Indian Institute of Technology Kanpur (IITK), India.

As a part of flight test program a real flight database, through various maneuvers, have been generated for the designed unmanned configuration. A dedicated flight data acquisition system, capable of onboard logging and telemetry to ground station, has been used to record the flight data during these flight test experiments. In order to identify the systematic errors in the measurements, the generated flight data has been processed through data compatibility check.

It is observed from the flight path reconstruction that the obtained biases are negligible and the scale factors are almost close to unity. The linear aerodynamic model along with maximum likelihood and least-square methods have been used to perform the parameter estimation from the obtained compatible flight data. The lower values of Cramer-Rao bounds obtained for various parameters has shown significant confidence in the estimated parameters using maximum likelihood method. In order to validate the aerodynamic model used and to increase the confidence in the estimated parameters a proof-of-match exercise has been carried out.

The entire work presented is original and all the experiments have been carried out in Flight laboratory of IITK.

The purpose of this paper is to provide a systematic overview with guidelines how to use partial least squares (PLS) path modeling in longitudinal studies. Practical examples from a study of the acceptance of battery electric vehicles (BEVs) in corporate fleets are used for demonstration purposes.

In this study, data at three points in time were collected: before the initial use of a BEV, after three and after six months of extensive usage of BEVs.

Three different models are identified depending on the research objective and on the data basis. Multigroup analyses are suggested to test the difference between the path coefficients of latent variables at different points in time. Limitations for the use of repeated cross-sectional data have to be observed.

Academics and practitioners will benefit from this paper by receiving an overview of the different PLS path models in longitudinal studies. A decision-tree enables them to make a choice regarding the most appropriate model and suggests a sequence of complementary analyses. So far, there is a lack of a tutorial type paper delivering such guidance.