Content uploaded by Craig J. Russell
Author content
All content in this area was uploaded by Craig J. Russell on Jan 28, 2016
Content may be subject to copyright.
A preview of the PDF is not available
In pursuit of moderation: Nine common errors and their solutions
Abstract
One result of the increasing sophistication and complexity of MIS theory and research is the number of studies hypothesizing and testing for moderation effects. A review of the MIS and broader management literatures suggests researchers investigating moderated relationships often commit one or more errors falling into three broad categories: inappropriate use or interpretation of statistics, misalignment of research design with phenomena of interest, and measurement or scaling issues. Examples of nine common errors are presented. Commission of these errors is expected to yield literatures characterized by mixed results at best, and thoroughly erroneous results at worse. Procedures representing examples of best practice and reporting guidelines are provided to help MIS investigators avoid or minimize these errors.
... In the second model, we estimate the effect of the CPI as a proxy for corruption, which we found to be non-significant. According to (Carte and Russell, 2003;Dawson, 2014;Borau et al., 2015), the significance of the link between the moderating variable and the dependent variable is not a necessary condition to attest to the existence of a moderating effect. In particular, we notice that the effect of ALMP expenditure is more attenuated than that found in Model 1. ...
This research is focused on investigating the efficiency of Active Labor Market Policies (ALMP) expenditures of Organisation for Economic Co-operation and Development (OECD) countries including institutional quality of governance. The emphasis is on the quality of governance that affects the supply and demand of employment and the ability of governments to master public spending. Using the general method of moments (GMM) for sixteen OECD countries over the 2005/2018 period, we show that Corruption Perception Index (CPI) reduced significantly the efficiency of ALMP on unemployment rate. This result is found consistent using another corruption proxy, Corruption Kaufmann Index (CKI). Moreover, our estimation suggest that the effectiveness of governance reinforces the effectiveness of ALMP.
... The results reveal that price consciousness moderates the effects of perceived value and user satisfaction on intention to pay (β = − 0.06, p < 0.01; β = − 0.10, p < 0.001, respectively). According to Carte and Russell (2003), an F-value significantly greater than 1 indicates that adding the moderator, price consciousness in our case, has a significant effect. We note an F-value of 26.20 in the moderation analysis. ...
A freemium strategy represents a popular business model for cloud-empowered services, in which a vendor gives away basic services for free and seeks to monetize premium services. The economic viability of freemium models depends on the vendor’s ability to convert free service users into paying customers. Different from prior studies that mainly examine users’ intention to pay, this study examines both users’ intentions to pay and their actual purchase behaviors by extending the IS success model to propose a software as a service (SaaS) success model. Our proposed model emphasizes two freemium-specific, user-oriented factors: (1) the fit between perceived value of provided services and user needs of premium services as a contingent factor, and (2) price consciousness resulted from free-mentality of SaaS as a moderating factor to understand how individual evaluations of provided services might lead to purchases of premium services. We empirically examine the proposed SaaS success model with data from 638 active SaaS users. The results indicate that SaaS users’ intentions to pay and actual purchase behaviors are shaped by their perceived value of and satisfaction with the provided services, moderated by price consciousness. Furthermore, the fit between perceived value and user needs has an important, contingent role in influencing users’ intentions to pay and purchase behaviors.
... The study conducted a comparison of the coefficients associated with individual paths within the two structural models to examine our hypotheses. The initial examination was directed towards gauging whether users of Facebook and Instagram perceived the latent constructs similarly (Carte & Russell, 2003). Analysis of factor loadings and cross-factor loadings ( Table 5) from measurements indicated a high degree of similarity in loading patterns, thereby permitting comparisons between paths across different contexts. ...
In the past few years, there has been a significant focus on investigating the impact of social media engagement. Although previous studies recognize the importance of social media engagement, there is still a lack of in-depth analysis regarding the specific influences of likes, comments, and shares on consumer behaviour, especially concerning how these engagements affect different platforms. This study aims to examine the relationship between social media engagement metrics (likes, comments, and shares) and consumer purchase intention on two prominent platforms, Facebook and Instagram. By adopting a cross-sectional design, this study employed purposive sampling to gather data from Malaysian university students who have engaged in purchasing activities on social media platforms such as Facebook and Instagram. The analysis was conducted using Partial Least Squares Structural Equation Modeling (PLS-SEM) based on 265 completed responses, comprising 131 from Facebook users and 134 from Instagram users. The results reveal a significant positive relationship between likes and shares with consumer purchase intention on both Facebook and Instagram. However, comments were not found to be associated with consumer purchase intention on either platform. Furthermore, the study conducted a comparative analysis between Facebook and Instagram and found no significant differences in the relationships between engagement metrics and purchase intention. These findings have practical implications for marketers, emphasizing the importance of fostering engagement through likes and shares to drive consumer purchasing behaviour on social media platforms.
... We follow the multi-step approach to reporting the results [114]. In the controls-only model (i.e., Table 3), confirmation bias (p<0.05) and ability (p<0.001) both had significant positive main effects, and benevolence (p<0.05) had a significant negative main effect, but integrity was not significant (p=0.071). ...
... When the moderator is from the nomological net, we recommend focusing on how the moderator impacts the dependent variable. After all, moderators also impact dependent variables, and when a moderator is present, the main relationship alone is not so meaningful (Carte and Russell, 2003). For example, a study may examine how the relationship between spatial distances and coordination delay among employees is moderated by communication technologies (e.g., Cummings et al., 2009). ...
Intradisciplinary research refers to research that integrates ideas often associated with different research domains in a discipline. Such cross-fertilization leverages abundant diversity present in the IS discipline to tackle increasingly complex IS problems and grand challenges. Despite its importance and recent attention, a concerted, sustained effort toward intradisciplinary research is lagging. A fundamental issue we see is a lack of an elaborate IS research map that effectively shows similarities and differences among research domains and demonstrates types of ideas that may travel and integrate into different domains. We thus aim to propose an elaborate IS research map and compile research elements that can be tried and combined across research domains. To do so, we utilize two IS classics (i.e., IT views and IS phenomena), identify their complementarity, and inter-weave the two disparate ways of portraying the IS research field. The resultant view-phenomenon (VP) matrix specifies research domains based on two consistent, comprehensive criteria and helps researchers discern similarities and differences among research domains more effectively. The VP matrix also sheds light on a variety of research elements that can flow across research domains. The VP matrix along with the research elements together facilitate intradisciplinary efforts and, more broadly, help the IS discipline to prosper. The VP matrix is particularly helpful for doctoral students and young scholars.
... These values are greater than 0, meaning the current study model has good relevance [73]. Carte and Russell [74] stated that there would be no effect size if the F 2 value is below 0.02. The size effect will be small for the F 2 value, which ranges from 0.02 to 0.15, but if the F 2 value is within 0.15-0.35, ...
Innovation is a vital element of an organization's success. Companies must implement efficient, innovative plans and policies to consistently develop novel innovations. Based on the theory of planned behavior, the current framework formulated the idea of entrepreneurial education, entrepreneurial mindset, and technological knowledge with entrepreneurial intention in the healthcare industry. Further, this concept comprises the mediating role of innovative work behavior and the moderating effect of entrepreneurial motivation between constructs. The study adopted a quantitative method and collected data from 529 employees working in healthcare organizations (e.g., medical services, manufacturing medical equipment, and drugs/Pharmaceutical companies). The study's findings revealed that entrepreneurial education, entrepreneurial mindset, and technological knowledge positively and significantly affect entrepreneurial intention. The mediation results show that innovative work behavior significantly affects other constructs, and entrepreneurial motivation has a significant moderating effect. The study highlighted the notable transformations in the healthcare industry by proposing the need for relevant entrepreneurial education, traits, and characteristics in accelerating business activities. The study found a link between entrepreneurial education, mindset, technological knowledge, and entrepreneurial intention. This study is helpful for managers and practitioners in executing effective strategies for innovation in healthcare companies.
... However, we could not confirm any influence of social support and social pressure on player performance (i.e., H3a, H6a, and H6b were not supported). Following Carte and Russell (2003), we conducted hierarchical regression in two models shown in Appendix G (Tables G1 and G2). Model I includes the CVs, and Model II adds four predictors. ...
Social factors play a critical role in motivating player participation and commitment to online multiplayer games. Many popular mobile massively multiplayer online games (MMOGs) adopt social network embeddedness (SNE) functions to optimise players' social play experience. SNE changes the traditional pattern of MMOG social play by porting acquaintance relationships (e.g., Facebook friends) from social networking sites to the virtual game world. However, little understanding exists on how SNE impacts mobile MMOG players' game participation results such as play performance and play frequency. Drawing on the affordance framework and social capital literature, this research proposes a theoretical model that integrates the factors of SNE technology affordance (identity transparency and information transparency), players' social experience (social interaction, social support, shared vision, and social pressure), plus affordance effects (play performance and play frequency). The model was validated through a longitudinal field study, in which both subjective and objective data were collected from Game for Peace players. Our findings indicate that identity transparency and information transparency positively correlate with social interaction, social support, shared vision, and social pressure, which, taken together, significantly affect play frequency. The results also show that social interaction and shared vision positively impact players' play performance. The study enhances the theoretical understanding of social relationships in players' game participation results from the SNE aspect. Finally, we lend insights on how game operators can improve player game experience and stickiness.
... Moderation analyses are based on linear regressions with interaction terms. We should note that the moderation analyses were not part of our preregistered hypotheses, and we expected that a larger sample size would be needed to detect significant moderation effects (Aiken, West, and Reno 1991;Carte and Russell 2003). ...
Despite a steep rise in income inequality, Americans’ preferences for redistribution have remained stagnant. Previous research suggests that egalitarian preferences are rooted in stable institutional and cultural contexts, but can change with exposure to information. We investigate the role of understandings of the link between income and psychological well-being in shaping policy preferences. In a large, preregistered online experiment (N=2,751), we examined the effects of three common themes in research: Money Prevents Unhappiness, Money Provides Happiness, and Money Doesn’t Matter. Results show that learning that Money Prevents Unhappiness (versus the other two themes) increased egalitarian preferences. Effects were moderated by political ideology, income, and social class, but not by race. We discuss implications in light of the current cultural discourse about happiness, which often privileges non-income causes and positive emotions.
The elderly face elevated mortality risk due to rising temperature. Previous assessments of temperature-related mortality, however, lack a comprehensive analysis of distinct impacts of temperature change across different timescales and characteristics. Using a longitudinal survey of 27,233 elderly Chinese citizens from 2005 to 2018, we establish connections between rising temperatures, temperature variability, and extreme heat with increased mortality risk, assessed through four annual metrics that combine temperature and humidity. The intensity and prolonged duration of extreme heat are found to have the greatest impact on mortality risk. Furthermore, by identifying heterogeneous impacts based on location, sex, age, obesity, income, and diet, we reveal the pathways through which temperature metrics are likely to influence mortality risk. Our study highlights the compound effects of rising temperatures for elderly populations, and it could be expanded to other countries and regions experiencing similar challenges due to an aging population experiencing warming conditions.
The traditional methodology for assessing moderator variables (hierarchical multiple regression analysis) is examined. Possible drawbacks of this technique for corroborating psychological theories (cf. Busemeyer & Jones, 1983), are illustrated empirically on the basis of an analysis of 400,000 subjects. This article tested a well-known (and currently popular) substantive hypothesis: A synergistic relation exists between mathematical ability and spatial visualization in the prediction and development of sophisticated levels of advanced mathematics. Using the traditional methodology, this hypothesis was confirmed; however, on further analysis, using a more systematic approach, it was demonstrated that this finding was spurious. Suggestions are offered for modifying the traditional methodology used for assessing moderator effects (for both applied and theoretical purposes). These amount to ways for minimizing Type I and Type II errors.
A discriminant analysis was performed on a sample (n = 114) of salaried professionals to distinguish work patterns among three age groups corresponding to early, middle and late career stages. Three distinct profiles were generated by the two significant discriminant functions which emerged. The three career stage groups were given identifying labels: ‘finding a niche’, ‘digging in’ and ‘entrenched’. The findings support a life-cycle hypothesis of ageing and work patterns, although there is some substantiation for the cohort and ageing hypotheses in the literature. Scenarios for each of the three career stages are provided along with implications for improving the management of age-specific professional human resources.
Although the concept of moderator variables has been used extensively in marketing-related studies, much confusion persists as to how they are defined and identified. To alleviate this confusion, the authors present a typology of moderator variables with a framework for identifying their presence and type. Simulated data are used to illustrate and validate the proposed framework.
Because the notion that various individual, task, or organizational-level variables moderate the relationships between leader behaviors and subordinate criterion variables is a fundamental assumption of most situational approaches to leadership, it is not surprising that a number of studies have been conducted in order to test these effects. Unfortunately, no systematic assessment of this research has been conducted to determine: (a) the nature of the tests for moderation that have been used, or (b) the degree to which the findings actually support the hypothesis that the effects of leader behaviors are moderated by situational variables. For the purposes of this review, we identified studies designed to test the moderators specified by two related theories of leadership: House’s (1971) Path-Goal Approach, and Kerr and Jermier’s (1978) Substitutes for Leadership Model. We then examined the methods used by researchers who tested for the moderating effects predicted by these models, the percentage of moderating effects actually found, and the nature of the moderating effects identified, The results indicate that: (1) although the percentage of moderators identified is clearly better than chance; (2) many of the tests conducted to identify moderating effects in this leadership area are inappropriate; (3) most of the results reported in this domain have not been replicated; and (4) the information reported by many of the authors is incomplete. The implications of the results of the review for both researchers and practicing managers are then discussed.
This article applies the concepts of alpha, beta, and gamma changes to test whether the implementation of a new office information system with networking capabilities changes the way organizational members conceptualize office work. The traditional approach (t-test) was used to measure alpha change and indicated little change in how effectively the respondents felt they performed eight generic office activities before implementation (T1) and nine months after implementation (T2). However, considerable change was detected between effectiveness reported at T1 and a retrospective assessment of T1 effectiveness reported at T2 (called “then” assessments). Strong change was also detected between “then” assessments and T2 effectiveness reported at T2, indicating beta change. Multiple hierarchical tests showed that most of the change was actually gamma change; the T2 and the “then” factor structures and covariances differed significantly. This study supports propositions that using computers to accomplish organizational work may be associated with different conceptualizations of work, which may create ambiguity and uncertainty if training and management policies do not respond appropriately. Finally, this study provides an expanded version of a prior solution to detecting alpha, beta, and gamma changes.
This research asks whether codes of ethics affect computer abuse judgments and intentions of information systems (IS) employees. Codes of ethics examined include both company codes of ethics and those written specifically to deal with IS issues. In addition, since the intent of codes of ethics is to clarify responsibility and deter unethical behavior, both the psychological trait of responsibility denial and its moderating effect on codes was studied. While company codes did not affect the computer abuse judgments and intentions of all IS personnel, they did affect those IS personnel who tend to deny responsibility, thus suggesting that company codes may clarify responsibility and reduce rationalizations for some people. Unlike company codes, IS-specific codes of ethics had a direct effect on computer sabotage judgments and intentions, but had no differential effect on those high in responsibility denial. Finally, responsibility denial was directly related to all computer abuse judgments and intentions studied. Overall, codes had little effect on computer abuse judgments and intentions relative to the psychological trait of responsibility denial.
Statistical power is a topic of importance to any researcher using statistical inference testing. Studies with low levels of statistical power usually result in inconclusive findings, even though the researcher may have expended much time and effort gathering the data for analysis. A survey of the statistical power of articles employing statistical inference testing published in leading MIS journals shows that their statistical power is, on average, substantially below accepted norms. The consequence of this low power is that MIS researchers typically have a 40 percent chance of not detecting the phenomenon under study, even though it, in fact, may exist.Fortunately, there are several techniques, beyond expanding the sample size (which often may be impossible) that researchers can use to improve the power of their studies. Some are as easy as using a different but more powerful statistical test, while others require developing more elaborate sampling plans or a more careful construction of the research design. Attention tot he statistical power of a study is one key ingredient in assuring the success of the study. This article should serve as a useful guide for MIS researcher sin the planning, execution, and interpretation of inferential statistical analyses.
Alpha, beta, and gamma change concerning student attitudes toward a college course were assessed before and after the first examination in that course for an ex perimental and control group. Three methodologies were used to assess change. Those proposed by Ter borg, Howard, and Maxwell (1980) and Schmitt (1982) produced reasonably similar conclusions con cerning change, while the methodology suggested by Zmud and Armenakis (1978) produced relatively dif ferent conclusions. The relative advantages and limita tions of the procedures are discussed. The major con clusion is that much additional use and comparison of these methodologies for assessing change is necessary before researchers or practitioners can interpret the practical significance of beta and gamma change or the relative utility of various approaches to the measure ment of beta and gamma change.
Having estimated a linear regression with p coefficients, one may wish to test whether m additional observations belong to the same regression. This paper presents systematically the tests involved, relates the prediction interval (for m = 1) and the analysis of covariance (for m > p) within the framework of general linear hypothesis (for any m), and extends the results to testing the equality between subsets of coefficients.