Variance-based SEM, also known under the term partial least squares (PLS) analysis, is an approach that has gained increasing interest among marketing researchers in recent years. During the last 25 years, more than 30 articles have been published in leading marketing journals that have applied this approach instead of the more traditional alternative of covariance-based SEM (CBSEM). However, although an analysis of these previous publications shows that there seems to be at least an implicit agreement about the factors that should drive the choice between PLS analysis and CBSEM, no research has until now empirically compared the performance of these approaches given a set of different conditions. Our study addresses this open question by conducting a large-scale Monte-Carlo simulation. We show that justifying the choice of PLS due to a lack of assumptions regarding indicator distribution and measurement scale is often inappropriate, as CBSEM proves extremely robust with respect to violations of its underlying distributional assumptions. Additionally, CBSEM clearly outperforms PLS in terms of parameter consistency and is preferable in terms of parameter accuracy as long as the sample size exceeds a certain threshold (250 observations). Nevertheless, PLS analysis should be preferred when the emphasis is on prediction and theory development, as the statistical power of PLS is always larger than or equal to that of CBSEM; already, 100 observations can be sufficient to achieve acceptable levels of statistical power given a certain quality of the measurement model.

Selection Procedure

The Jan-Benedict E.M. Steenkamp Award for Long-Term Impact is given annually to papers published in IJRM in recognition of their exceptional contributions to academic marketing research by demonstrating long-term impact.

A 4-member Award Committee, formed by the IJRM Editor-in-Chief (PK Kannan) and Co-Editor (Iris W. Hung), managed the nomination and selection procedure. For 2020, the committee was composed of Praveen Kopalle, chairperson (Dartmouth College, USA), Ujwal Kayande (Melbourne Business School, Australia), Sharon Ng (Nanyang Technological University, Singapore), and Alina Sorescu (Texas A&M University, USA).

The selection procedure for this award is as follows:

1. All papers published in IJRM 10 to 15 years prior to the year the award is being presented are eligible. Thus, for the 2021 Jan-Benedict E.M. Steenkamp Award, 181 papers published in the years 2006 through 2011 were eligible. (Papers published within this time frame that have already won this award or are (co)authored by Jan-Benedict Steenkamp and/or by the current IJRM Editor-in-Chief were not eligible.) Nominations were invited from EMAC members and IJRM Editorial Board members. This year, the Award Committee received nominations for 76 papers. These nominated papers comprised the first ballot for the first round of voting open only to the Members of the IJRM Editorial Board (who could vote for up to 5 papers; self-voting was not allowed.) The eleven (11) papers that received the most votes in the first round made up the ballot for the second and final round of voting in which the Editorial Board could choose only 1 paper. (This year there were 11 instead of 10 papers due to a tie for the 10^th place.)

2. After receiving the votes, the Award Committee deliberated on the winning paper guided by the following criteria: (1) the votes received from the IJRM Editorial Board from the two rounds of voting, (2) its ISI and Google Scholar citations, and (3) its quality, as assessed by the committee's in-depth reading. There can be two winners in exceptional cases (not more than once every 3 years on average).

Statement from the Award Committee

During the past two decades, there has been a renewed interest among researchers to use partial least squares analysis (PLS). However, prior research has not evaluated the relative performance of covariance- based structural equation modeling (CBSEM) and PLS approaches per a set of key characteristics such as sample size, number of indicators per construct, distributional assumptions etc. An interesting and important question is whether we can arrive at a set of rules that researchers may use in their choice of CBSEM versus PLS. The paper by Reinartz, Haenlein, and Henseler explores this issue and examines whether each of the approaches converge to a proper solution, the degree of parameter accuracy between the approaches, the relative importance of the different design factors on parameter accuracy, statistical power etc.

The authors conduct a set of Monte Carlo simulations based on 240 scenarios defined by a full factorial design of four design factors. The results suggest that the statistical power of PLS is always larger than or equal to that of maximum likelihood-based CBSEM. Further, their simulations show that PLS can be a very good methodological choice if sample size is low. Finally, their results indicate that CBSEM is actually extremely robust to violations of its underlying distributional assumptions.

This paper led by a clear margin by receiving the most votes of the IJRM Editorial Board and was approved unanimously by the award committee. In addition, the evidence of its impact is highlighted by 1039 Web of Science and 2320 Google Scholar citations, which makes it the best cited paper among all the initial round award nominees and shows its consistent use by scholars to decide which method to use.

We congratulate the authors for receiving this award.

The 2021 Award Committee:

Praveen Kopalle, chairperson, Ujwal Kayande, Sharon Ng, and Alina Sorescu

11 Shortlisted papers:

•

Will the frog change into a prince? Predicting future customer profitability. Roland T. Rust, V. Kumar, Rajkumar Venkatesan. Pages 281-294, Vol 28 (4)

•

Agent-Based Modeling in Marketing: Guidelines for Rigor. William Rand and Roland T. Rust. Pages 181-193, Vol 28 (3)

•

Generalizations on consumer innovation adoption: A meta-analysis on drivers of intention and behavior. Joep W.C. Arts, Ruud T. Frambach, Tammo H.A. Bijmolt. Pages 134-144, Vol 28 (2)

•

Innovation diffusion and new product growth models: A critical review and research directions. Renana Peres, Eitan Muller, Vijay Mahajan. Pages 91-106, Vol 27 (2)

•

The chilling effects of network externalities. Jacob Goldenberg, Barak Libai, Eitan Muller. Pages 4-15, Vol 27 (1)

•

An empirical comparison of the efficacy of covariance-based and variance-based SEM. Werner Reinartz, Michael Haenlein, Jörg Henseler. Pages 332-344, Vol 26 (4)

•

A new measure of brand personality. Maggie Geuens, Bert Weijters, Kristof De Wulf. Pages 97-107, Vol 26 (2)

•

Measuring the impact of positive and negative word of mouth on brand purchase probability. Robert East, Kathy Hammond, Wendy Lomax. Pages 215–224, Vol 25 (3)

•

Reaping relational rewards from corporate social responsibility: The role of competitive positioning. Shuili Du, C.B. Bhattacharya, Sankar Sen. Pages 224-241, Vol 24 (3)

•

The NPV of bad news. Jacob Goldenberg, Barak Libai, Sarit Moldovan, Eitan Muller. Pages 186–200, Vol 24 (3)

•

Antecedents and purchase consequences of customer participation in small group brand communities. Richard P. Bagozzi, Utpal M. Dholakia. Pages 45-61, Vol 23 (1)

基于方差的 SEM，也称为偏最小二乘 (PLS) 分析，是近年来市场研究人员越来越感兴趣的一种方法。在过去的 25 年中，在领先的营销期刊上发表了 30 多篇文章，这些期刊应用了这种方法，而不是更传统的基于协方差的 SEM (CBSEM) 替代方法。然而，尽管对这些先前出版物的分析表明，对于应该推动 PLS 分析和 CBSEM 之间的选择的因素似乎至少有一个隐含的同意，但迄今为止，没有研究根据经验比较这些方法的性能，给定一组不同的条件。我们的研究通过进行大规模的蒙特卡罗模拟来解决这个悬而未决的问题。我们表明，由于缺乏关于指标分布和测量尺度的假设来证明选择 PLS 通常是不合适的，因为 CBSEM 证明在违反其基本分布假设方面非常稳健。此外，CBSEM 在参数一致性方面明显优于 PLS，并且只要样本量超过某个阈值（250 个观察值），就参数准确性而言更可取。然而，当强调预测和理论发展时，应该首选 PLS 分析，因为 PLS 的统计能力总是大于或等于 CBSEM；考虑到一定的测量模型质量，100 个观测值已经足以达到可接受的统计功效水平。

选择程序

Jan-Benedict EM Steenkamp 长期影响奖每年颁发给在 IJRM 上发表的论文，以表彰他们通过展示长期影响对学术营销研究做出的杰出贡献。

由IJRM主编 (PK Kannan) 和联合编辑 (Iris W. Hung)组成的 4 人奖项委员会负责管理提名和遴选程序。2020年委员会由主席Praveen Kopalle（美国达特茅斯学院）、Ujwal Kayande（澳大利亚墨尔本商学院）、Sharon Ng（新加坡南洋理工大学）和Alina Sorescu（美国德克萨斯农工大学）组成。

该奖项的评选程序如下：

1.获奖前 10 至 15 年在IJRM 上发表的所有论文均有资格。因此，对于 2021 年Jan-Benedict EM Steenkamp 奖，2006 年至 2011 年发表的181篇论文符合条件。（论文那些已经获得这一奖项或者是（共此时限内公布），作者：扬笃Steenkamp和/或由当前IJRM主编，主编是不符合条件的。）提名从EMAC成员和IJRM编辑邀请董事会成员。今年，奖项委员会收到了76项提名文件。这些被提名的论文包括第一轮投票的第一轮投票，仅对 IJRM 编辑委员会成员开放（他们最多可以对 5 篇论文进行投票；不允许自行投票。）获得投票的十一 (11) 篇论文第一轮的大多数选票构成了第二轮也是最后一轮投票的选票，其中编辑委员会只能选择 1 篇论文。（今年有11，而不是10篇论文因并列的10^个地方。）

2. 收到选票后，评奖委员会按照以下标准对获奖论文进行了审议：（1）IJRM 编辑委员会在两轮投票中获得的票数，（2）其 ISI 和 Google Scholar 的引用情况，以及(3) 它的质量，由委员会的深入阅读评估。在特殊情况下可以有两名获胜者（平均每 3 年不超过一次）。

颁奖委员会的声明

在过去的二十年中，研究人员重新对使用偏最小二乘分析 (PLS) 产生了兴趣。然而，先前的研究没有评估基于协方差的结构方程建模 (CBSEM) 和 PLS 方法的相对性能，例如样本大小、每个构造的指标数量、分布假设等。一个有趣且重要的问题是我们是否可以得出一组规则，研究人员在选择 CBSEM 与 PLS 时可能会使用这些规则。通过本文赖纳茨，Haenlein和Henseler 探讨这个问题并检查每种方法是否收敛到适当的解决方案、方法之间的参数准确度、不同设计因素对参数准确度、统计功效等的相对重要性。

作者根据由四个设计因子的全因子设计定义的 240 个场景进行了一组蒙特卡罗模拟。结果表明，PLS 的统计功效总是大于或等于基于最大似然的 CBSEM。此外，他们的模拟表明，如果样本量很小，PLS 可能是一个非常好的方法选择。最后，他们的结果表明 CBSEM 实际上对于违反其基本分布假设非常稳健。

这篇论文获得了IJRM编辑委员会的最多选票，以明显的优势领先，并获得了奖项委员会的一致批准。此外，1039 次 Web of Science 和 2320 次 Google Scholar 引用突出了其影响的证据，这使其成为所有首轮奖项提名者中被引用次数最多的论文，并表明学者们一致使用它来决定使用哪种方法。

我们祝贺作者获得此奖项。

2021年奖项委员会：

Praveen Kopalle，主席，Ujwal Kayande、Sharon Ng 和 Alina Sorescu

11篇入围论文：

•

青蛙会变成王子吗？预测未来客户的盈利能力。Roland T. Rust、V. Kumar、Rajkumar Venkatesan。第 281-294 页，第 28 卷 (4)

•

营销中基于代理的建模：严谨性指南。William Rand 和 Roland T. Rust。第 181-193 页，第 28 卷 (3)

•

消费者创新采用的概括：意图和行为驱动因素的元分析。Joep WC Arts、Ruud T. Frambach、Tammo HA Bijmolt。第 134-144 页，第 28 卷 (2)

•

创新扩散和新产品增长模型：批判性审查和研究方向。雷纳娜·佩雷斯、艾坦·穆勒、维杰·马哈詹 第 91-106 页，第 27 卷 (2)

•

网络外部性的寒蝉效应。雅各布·戈登伯格、巴拉克·利白、艾坦·穆勒。第 4-15 页，第 27 卷 (1)

•

基于协方差和基于方差的 SEM 功效的实证比较。维尔纳·莱纳茨、迈克尔·海恩莱恩、约尔格·亨塞勒 第 332-344 页，第 26 卷 (4)

•

品牌个性的新衡量标准。玛吉·格恩斯、伯特·魏特斯、克里斯托夫·德·沃尔夫 第 97-107 页，第 26 卷 (2)

•

衡量正面和负面口碑对品牌购买概率的影响。罗伯特·伊斯特、凯西·哈蒙德、温迪·洛马克斯。第 215–224 页，第 25 卷 (3)

•

从企业社会责任中获得相关回报：竞争定位的作用。Shuili Du, CB Bhattacharya, Sankar Sen. Pages 224-241, Vol 24 (3)

•

坏消息的 NPV。雅各布·戈登伯格、巴拉克·利白、萨里特·摩尔多万、艾坦·穆勒。第 186–200 页，第 24 卷 (3)

•

客户参与小集团品牌社区的前因和购买后果。Richard P. Bagozzi，Utpal M. Dholakia。第 45-61 页，第 23 卷 (1)