Contributions published at Marketing and Market Research (René Algesheimer)

null

Media plays an important role to our everyday lives because we shape our opinion and make informed decisions by relying on information obtained from the media. Therefore the media is expected to provide fair and unbiased media coverage. However this appears far from reality. In the last years media consumers’ trust in media has decreased; consumers perceive the media as biased and believe that powerful individuals and organizations influence the media. Indeed, some scholars argue that increased competition, ownership concentration and conglomeratization in the media industry have greatly contributed to media bias. The aim of this literature-based thesis for one thing is to provide a general overview on the topic of media bias and its manifestations, and for the other thing, to explain the causes and consequences of media bias while discussing the subject within the context of a general institutional problem inherent in media industry (structural media bias).

System dynamics (SD), an approach to modeling and simulating complex systems, has repeatedly demonstrated its value in contributing to the understanding and solution of complex policy problems. Typical areas of system dynamics application include modeling of policy problems related to public health, energy and the environment, social welfare, sustainable development, and security. One of the main challenges in system dynamics is that, due to a high degree of interdependent model variables and nonlinear relationships, the detection of model levers, i.e. variables capable of effectively and efficiently controlling complex policy problems, is exceedingly demanding. So, notwithstanding the usefulness of system dynamics in the analysis of these problems, the solution identification process is far from straightforward and in most cases trial-and-error driven. To address this challenge, we propose to combine system dynamics with network controllability to facilitate the detection of model levers. In essence, a system dynamics model can be thought of a web of interrelated causal factors that are assumed giving rise to the complex policy problem under study. Due to its web similarity, the structure of a system dynamics model can be accurately described as a directed weighted network, making it accessible to algorithmic exploration using concepts from the fields of graph theory and network science. Referring to recent research on control principles of complex networks, model levers are found first by calculating the size of the minimum driver set of the system dynamics model (network), second by computing all existing minimum driver sets, and third by ranking minimum driver set variables according to their control centrality. Variables with a high control centrality should be of primary interest to policy-makers when designing new solutions to complex policy problems. We demonstrate the proposed multimethodological approach on the basis of the World Dynamics model, a classic example from the system dynamics literature.

null

null

null

Examining the complex role of formalization within a globalized context, we develop and test a new multi-level theory to address the positive and negative effects of formalization on organizational citizenship behavior (OCB). We review classic assumptions of culture-fit theories that argue that organizational practices are most effective if they match the larger technological and cultural context. We argue that there is a complex dynamic in that formalization has negative associations with OCB in national contexts that are relatively certain but positive associations with OCB in contexts of higher environmental uncertainty. A three-level hierarchical linear modeling analysis of data from 7,537 employees in 267 organizations across 17 countries supports our hypotheses. In nations with greater levels of uncertainty, formalization is positively associated with voice OCB. Effects on helping are more complex and appear masked by other culture-level dimensions. Our findings emphasize the need for further study of the dynamics between national contexts, organizational cultures and individual behavior.

null

In today's global and interconnected world, where people can operate from distance through the Internet or telecommunication, cheating behavior has become a crucial topic in the area of economic businesses. Organizations got highly vulnerable to fraudulent attacks through external and/or internal individuals. The aim of this thesis is to gain a deeper understanding on why “ordinary” people, like you and me, who have high moral views of themselves, engage in unethical behavior. The psychological mechanisms that lead to cheating behavior are analyzed on an individual level. After defining fraudulent and unethical behavior, past research in this area is reviewed and categorized in different research streams. The literature review gives an overview about the standard economic, psychological, behavior economic and the neuroscientific perspectives on cheating behavior. Subsequently, the mechanisms behind intentional and unintentional unethical behavior are examined. The thesis is finalized with some managerial implications on how to curb cheating behavior and with concluding remarks.

null

Many community detection algorithms have been developed to uncover the mesoscopic properties of complex networks. However how good an algorithm is, in terms of accuracy and computing time, remains still open. Testing algorithms on real-world network has certain restrictions which made their insights potentially biased: the networks are usually small, and the underlying communities are not defined objectively. In this study, we employ the Lancichinetti-Fortunato-Radicchi benchmark graph to test eight state-of-the-art algorithms. We quantify the accuracy using complementary measures and algorithms' computing time. Based on simple network properties and the aforementioned results, we provide guidelines that help to choose the most adequate community detection algorithm for a given network. Moreover, these rules allow uncovering limitations in the use of specific algorithms given macroscopic network properties. Our contribution is threefold: firstly, we provide actual techniques to determine which is the most suited algorithm in most circumstances based on observable properties of the network under consideration. Secondly, we use the mixing parameter as an easily measurable indicator of finding the ranges of reliability of the different algorithms. Finally, we study the dependency with network size focusing on both the algorithm's predicting power and the effective computing time.

The measurement of values using the 21-items Portrait Value Questionnaire is included in the European Social Survey (ESS). These data cover population samples of European countries and offer a unique opportunity to compare value priorities across countries. A necessary precondition for meaningful comparisons is the establishment of measurement invariance (MI). Unfortunately, previous studies could not establish MI of values across ESS countries. Studies which tested for MI used exact invariance approach which is demanding. However, recent literature suggests using a more liberal approximate approach. In the current study we applied this new method that uses a Bayesian framework. We tested for the approximate MI properties of values across 15 countries participating in six ESS rounds (total N = 274,447). It turned out that approximate MI was given in almost all ESS rounds for self-enhancement (8 countries), self-transcendence (12 countries), and conservation (10 countries).

Personal value priorities (the circular model of Schwartz) and personality traits (the Big Five of Costa and McCrae) belong to the most important individual psychological characteristics in cross-cultural research. Empirically supported relations between these constructs, although theoretically meaningful, have not led so far to the development of a comprehensive model of personality that includes both values and traits. The recently proposed Circumplex of Personality Metatraits (CPM; Strus, Cieciuch, Rowiński, 2014), built on two higher-order factors of personality (Alpha and Beta), suggests such an integrative model. The CPM is inspired by the idea of value circularity and considers recent criticisms of the Big Five model, thus suggesting Big Two rather than Big Five personality traits for better replicability in cross-cultural research. The empirical verification of the integration of values and traits in the CPM model is presented in a series of studies conducted in Poland (N > 2,500).

Values are relatively stable in adults, but less so in children. However, within a single time period, individuals should be consistent (i.e., not vary greatly) in their value choices. That is, a value that is more important than other values should be consistently preferred, regardless of which other values there are to choose from. To examine value consistency, we administered an animated BWS values instrument to children (5-12) in Australia (N=829) and Poland (N=555). This instrument asks respondents to choose the value-items they most (and least) like from 21 sets of 5 value-items. Over 70% of children were very consistent in the choice of value they like most (chosen at least 4 of the 5 times it appeared) and approximately 80% of children were very consistent in the choice of value they like least. We found that younger children exhibited less consistency.

A special module about attitudes toward immigration and threat due to immigration was implemented in the 7th Round of European Social Survey (ESS). In our project we set two goals. The first one was to establish in a theory-driven way latent variables based on items included in the module. These latent variables can be used by researchers in their substantive work on immigration using the ESS. The second goal was to test for measurement invariance of these scales across 15 ESS countries. We proposed the four following latent variables: allowing for immigrants belonging to different ethnic groups than the majority population into the country; qualification for entry; and two types of threat due to immigrants – realistic and symbolic. First, we tested each latent variable in each country separately in single Confirmatory Factor Analyses (CFAs). Next, we tested for measurement invariance of each latent variable using multigroup Confirmatory Factor Analysis (MGCFA). We differentiated between three levels of measurement invariance, configural, metric and scalar, and we applied two approaches: an exact and an approximate measurement invariance approach. If full or partial exact measurement invariance could not be established, we tested whether approximate invariance was given. Configural and metric invariance was supported for all constructs across most countries. Unfortunately scalar invariance was supported for the latent variables only across a subset of countries. The subset of countries where approximate scalar invariance was established was larger than the subset of countries for which exact measurement invariance could be established.

null

null

null

Today valuing customers is essential any firm and enables marketers to identify key customers. Customer lifetime value (CLV) is a central metric for valuing customers. It describes the long-term economic value of customers and gives managers an idea of how customers will evolve over time. Among the various approaches to assess CLV in non-contractual business settings probabilistic customer attrition models stand out due to their ability to simultaneously forecast a customer’s actual lifetime and future transactions. However, empirical evidence suggests that standard probabilistic customer attrition models do not outperform basic management heuristics. One of the main reasons for the limited predictive performance of the Pareto/NBD model is the assumption of independence of observed external determinants. Regularity patterns and non-random direct marketing activities are the two broad categories of contextual factors identified in previous literature to affect the predictive performance of latent probabilistic customer attrition models. Time-varying covariates would allow to control for these effects. Recently an extension to include time-varying covariates for continuous non-contractual business setting non has been presented. However an implementation of time-varying covariates for the continuous non-contractual business setting is not available. In this paper we propose a latent attrition model that allows for the first time to model time-varying contextual factors in continuous non-contractual settings. Our approach is an extension of the standard Pareto/NBD model. Our extension allows time-varying contextual factors to influence both, the purchase and the attrition process. Our findings have strong implications for both, marketing practice and research. Besides giving detailed recommendations on when to use which modeling approach, we also provide practical advices for estimating these models.

null