Empirical generalizations from brand extension research: How sure are we?

Echambadi, Rajagopal; Arroniz, Inigo; Reinartz, Werner; Lee, Junsoo

doi:10.1016/j.ijresmar.2006.02.002

Cited by 63 publications

(46 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Its standard error does not change either, as I show later, although its variance inflation factor (VIF) falls because inessential multicollinearity is alleviated. In contrast, the changes in all one-and two-way terms (such as the one-way simple effects in a QMMR testing a twoway interaction) represent systematic biases [1], which Lance [9] did not recognize in recommending the procedure. Instead, he attributed the changes in parameters of lower-order polynomial terms to improved estimation from reduced multicollinearity [2,9].…”

Section: Residual Centering (Orthogonalization)mentioning

confidence: 97%

“…Substituting d [xwz] for xwz in Equation (1), I re-estimate this raw regression as a residual centered regression, factoring out the variance in xwz that is shared with one-and two-way predictors:   …”

Section: Residual Centering (Orthogonalization)mentioning

confidence: 99%

“…Lance (Standardization, it should be noted, transforms one-way variables to become mean-centered, such that simple effects become main effects.) Standardization allows predictors to be compared to identify those with stronger effects 1 , although when non-arbitrary scaling metrics are used, only unstandardized estimates should be conducted, as Lance [9] recommends. 1 When predictors are arbitrarily scaled, Lance [9] recommended restandardizing the residually centered interaction, prior to its use in the final QMMR so that it will have a mean of zero and standard deviation of one, as do the related one-way standardized terms on which the interaction is based.…”

Section: Biased Interpretations From Residual Centeringmentioning

confidence: 99%

“…Progress in overcoming this predicament suffered a setback several years ago when it was proven that the common practice of mean centering in moderated regression cannot alleviate multicollinearity among variables comprising an interaction, but merely masks it. Residual centering (orthogonalizing) is unacceptable because it biases coefficients for predictors from which the interaction(s) derives, despite the non-biased coefficient for the highest-order polynomial interaction, thus precluding interpretation of moderator effects [1,2].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Could Sequential Residual Centering Resolve Low Sensitivity in Moderated Regression? Simulations and Cancer Symptom Clusters

Francoeur¹

2013

OJS

View full text Add to dashboard Cite

Multicollinearity constitutes shared variation among predictors that inflates standard errors of regression coefficients. Several years ago, it was proven that the common practice of mean centering in moderated regression cannot alleviate multicollinearity among variables comprising an interaction, but merely masks it. Residual centering (orthogonalizing) is unacceptable because it biases parameters for predictors from which the interaction derives, thus precluding interpretation of moderator effects. I propose and validate residual centering in sequential re-estimations of a moderated regression-sequential residual centering (SRC)-by revealing unbiased multicollinearity conditioning across the interaction and its related terms. Across simulations, SRC reduces variance inflation factors (VIF) regardless of distribution shape or pattern of regression coefficients across predictors. For any predictor, the reduced VIF is used to derive a lower standard error of its regression coefficient. A cancer sample illustrates SRC, which allows unbiased interpretations of symptom clusters. SRC can be applied efficiently to alleviate multicollinearity after data collection and shows promise for advancing synergistic frontiers of research.

show abstract

Section: Residual Centering (Orthogonalization)mentioning

confidence: 97%

Section: Residual Centering (Orthogonalization)mentioning

confidence: 99%

Section: Biased Interpretations From Residual Centeringmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Could Sequential Residual Centering Resolve Low Sensitivity in Moderated Regression? Simulations and Cancer Symptom Clusters

Francoeur¹

2013

OJS

View full text Add to dashboard Cite

show abstract

“…The importance of fit or congruence is well established in the marketing literature, particularly in terms of sponsorship (Fleck and Quester, 2007), celebrity endorsements (Lynch and Schuler, 1994) and brand extensions (Echambadi, Arroniz, Reinartz and Lee, 2006). Strong perceived fit has been…”

Section: H8 Perceived Leveraging Moderates the Above Relationships Smentioning

confidence: 99%

The Impact of Corporate Volunteering on CSR Image: A Consumer Perspective

et al. 2014

View full text Add to dashboard Cite

show abstract

Brand and its effect on user perception of search engine performance

Jansen

Zhang

Schultz

2009

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

In this research we investigate the effect of search engine brand on the evaluation of searching performance. Our research is motivated by the large amount of search traffic directed to a handful of Web search engines, even though many have similar interfaces and performance. We conducted a laboratory experiment with 32 participants using a 4 2 factorial design confounded in four blocks to measure the effect of four search engine brands (Google, MSN, Yahoo!, and a locally developed search engine) while controlling for the quality and presentation of search engine results. We found brand indeed played a role in the searching process. Brand effect varied in different domains. Users seemed to place a high degree of trust in major search engine brands; however, they were more engaged in the searching process when using lesser-known search engines. It appears that branding affects overall Web search at four stages: (a) search engine selection, (b) search engine results page evaluation, (c) individual link evaluation, and (d) evaluation of the landing page. We discuss the implications for search engine marketing and the design of empirical studies measuring search engine performance. IntroductionThere has been a rapid growth in the Web search engine market since its inception. Search engines continue to attract a large number of Web searchers and consistently rank as some of the heavily visited sites in the market in terms of the number of visitors (Alexa Internet Inc., 2008). There are numerous search engines on the Web (Wikipedia, 2008); however, only a handful dominates in terms of usage (Sullivan, 2008). From a technological point, this clustering of traffic is interesting because studies report that the performance of most of the major search engines is practically the same (c.f., Eastman & Jansen, 2003). Performance is typically defined as returning useful results and is measured by precision, which is the ratio of relevant documents to the total number of document returned at some point in the results listing. The interfaces of most search engines are also similar, namely, a text box, some verticals (i.e., tabs for searching the Web, Images, Audio, etc.), and a submit button. In studies of search engine interface usability, the results among various search engines have been similar (c.f., Wildemuth & Carter, 2002).Given the similarity in technology and interface design, why do only a small number of search engines dominate Web traffic? Certainly maintaining Web searching infrastructures on a large scale is expensive, but several well-financed companies have been unsuccessful. Do other elements affect the evaluation of a search engine's performance? Seeking the answers to these questions motivate our research.There could be many possible avenues to investigate. In a series of user studies concerning Web searching (Jansen, 2006;Jansen & McNeese, 2005), the participants completed pre-surveys concerning their Web searching habits. One question addressed which search engine the participant used and why. There were ma...

show abstract

Empirical generalizations from brand extension research: How sure are we?

Cited by 63 publications

References 19 publications

Could Sequential Residual Centering Resolve Low Sensitivity in Moderated Regression? Simulations and Cancer Symptom Clusters

Could Sequential Residual Centering Resolve Low Sensitivity in Moderated Regression? Simulations and Cancer Symptom Clusters

The Impact of Corporate Volunteering on CSR Image: A Consumer Perspective

Brand and its effect on user perception of search engine performance

Contact Info

Product

Resources

About