Journal of Marketing Analytics (2024) 12:1–5
https://doi.org/10.1057/s41270-023-00279-7
EDITORIAL
Advanced marketing analytics using partial least squares structural
equation modeling (PLS‑SEM)
Marko Sarstedt1,2 · Yide Liu3
Accepted: 24 November 2023 / Published online: 16 December 2023
© The Author(s) 2023
The use of partial least squares structural equation mod- 2023), predictive power (Shmueli et al. 2016), or endogene-
eling (PLS-SEM) as a means to analyze complex interre- ity assessment (Becker et al. 2022), and (3) the combined
lationships between latent variables and their indicators use of PLS-SEM with methods from machine learning (Ster-
(Hair et al. 2022; Lohmöller 1989; Wold 2005) has recently nad Zabukovšek et al. 2022), or other fields (e.g., Richter
surged—not only in fields that popularized the method such et al. 2020). This special issue brings together high-quality
as management information systems, marketing, and strate- papers that apply or advance these state-of-the-art extensions
gic management (e.g., Guenther et al. 2023; Hair et al. 2017; of the original PLS-SEM method, several of which were
Sarstedt et al. 2022), but in various other disciplines such as presented at the 2022 International Conference on Partial
agriculture, engineering, environmental sciences and ecol- Least Squares Structural Equation Modeling, held in Sep-
ogy, geography, and psychology. PLS-SEM’s methodologi- tember 6–9, 2022 at the Faculty of Economics and Business
cal features have clearly extended researchers’ capacities to Administration of the Babeș-Bolyai University, Cluj-Napoca,
better understand the complex interrelationships that consti- Romania.
tute the “black box” of a variety of attitudinal and behavioral The first two papers presented in this special issue intro-
theories as well as explain and predict unobservable phe- duce a new dimension to the analysis and interpretation of
nomena (e.g., Petter 2018; Petter and Hadavi 2021; Russo PLS-SEM results, which inherently follows a sufficiency
and Stol 2022). For example, articles recently published logic. According to this logic, the parameter estimates
in Journal of Marketing Analytics have used PLS-SEM to express the strength of the relationships between model
assess the effects of COVID-19-related risk on online shop- elements, most notably the constructs of interest. This is
ping behavior (Soares et al. 2023), the influence of storytell- reflected in authors’ interpretation of the results who typi-
ing on the consumer-brand relationship experience (Crespo cally use expressions such as “we find that attractiveness
et al. 2023), and the drivers of consumer trust in live stream- (…) and performance (…) have a particularly strong effect
ing platforms (Leong et al. 2023). Many of these studies on the cognitive corporate reputation construct, while
rely on advanced modeling and model assessment routines, attractiveness (…) and quality (…) are the most important
which have been developed over the last years (Hair et al. explanators for the affective corporate reputation dimension”
2024). Examples include (1) the analysis of complex model (Damberg et al. 2022, p. 9). While such findings are highly
relationships involving nonlinear effects (Basco et al. 2022), relevant for managerial practice, they imply that the absence
conditional mediating effects (Cheah et al. 2021), or higher- of a certain determinant such as attractiveness in Damberg
order models (Sarstedt et al. 2019), (2) the use of model et al.’s 2022 analysis can, in principle, be compensated by
evaluation routines for discriminant validity (Ringle et al. other determinants such as performance or quality. However,
intuition tells us that this is rarely the case.
Addressing this concern, Dul (2016a, b) proposed the
* Marko Sarstedt necessary condition analysis (NCA), which assumes that an
[email protected] outcome—or a certain level of an outcome—can only be
Yide Liu achieved if the necessary cause is in place or is at a certain
[email protected] level. Following this necessity logic, researchers may, for
1 example, conclude that achieving a certain level of repu-
Ludwig-Maximilians-University Munich, Munich, Germany
2
tation requires an attractiveness-level of 60%. The practi-
Babeș-Bolyai University, Cluj‑Napoca, Romania cal benefits of such conclusions are obvious as they offer
3
Macau University of Science and Technology, Macau, China
Vol.:(0123456789)
2 M. Sarstedt, Y. Liu
concrete managerial guidance as to which effort is needed to layers. This design feature of ANN is likely to produce a
produce a desired outcome. Not surprisingly, the NCA has mismatch between the network and the PLS path model,
recently attracted considerable attention (Dul et al. 2023) where no such requirement exists. Hence, when using the
and has also gained prominence as a means to enrich PLS- construct scores from a previous PLS-SEM analysis as input
SEM analyses. To derive necessary conditions, research- for the ANN estimation, researchers typically estimate the
ers applying PLS-SEM use the construct scores after algo- network layer-by-layer. When jointly applying PLS-SEM
rithm convergence as input for an NCA (Richter et al. 2020, and ANN, researchers also need to be aware that PLS-SEM
2022, 2023). The construct scores of the determinants and seeks to maximize in-sample prediction, while ANN analy-
the outcome are then contrasted in separate scatterplots ses focus on maximizing out-of-sample prediction. As such,
which include ceiling lines that characterize the maximum ANN-based robustness checks offer a valuable addition to
input–output relations (i.e., the highest possible outcome predictive power analyses on the grounds of techniques such
that can be achieved for a certain determinant construct as PLSpredict (Shmueli et al. 2016, 2019) or the cross-vali-
score). Each ceiling line’s slope and intercept facilitate quan- dated predictive ability test (Liengaard et al. 2019; Sharma
tifying the strength of the necessity condition via the neces- et al. 2023), while considering nonlinearities in the model
sity effect size d; bottleneck tables offer further insights into relations. Overall, ANN and PLS-SEM may be viewed as
the nature of the necessity conditions. Two articles in this synergistic in that the PLS path model offers a logical frame-
special issue also make use of the NCA, thereby showcasing work for the neural network, while the neural network can
the potentials of this multimethod approach. Damberg et al. furnish efficient parameter estimates for the structural model
(2023) complement their standard PLS-SEM analysis with relations. Mkedder et al.’s (2023) paper published in this
an NCA, showing that the relationships between corporate special issue considers this approach in order to evaluate
reputation, its determinants, and consumers’ (sustainable) the PLS path model’s robustness in the context of virtual
satisfaction are indeed necessary. Tiwari et al. (2023) run goods purchases.
an NCA on their technology acceptance model to further Another area of methodological development relates to
the understanding of factors that drive travel app use. For the specification, estimation, and validation of more com-
example, their analysis shows that the technology’s relative plex model relationships such as mediating effects. Mediat-
advantage and compatibility act as necessary conditions for ing effects assume that an antecedent construct impacts a tar-
perceived usefulness but not perceived enjoyment. get construct through a sequence of one or more intervening
While the NCA offers an alternative perspective on constructs, referred to as mediators. By analyzing sequences
the nature of model relationships, researchers have also of relationships, researchers shed light on the mechanisms
proposed means to substantiate the hypothesized model that underlie the assumed cause-effect relationships (Nitzl
structure. Inspired by machine learning, researchers have et al. 2016). When estimating such effects, researchers have
used the construct scores produced by the PLS-SEM algo- typically considered relatively simple mediation models with
rithm as input for artificial neural networks (ANN). ANN one or very few mediators whose effects they considered
simulate information processing through different layers of in isolation. With the increasing dissemination of the PLS-
neurons in the brain’s nervous system. This processing is SEM method, mediation analyses have also become more
characterized by activation functions, which dictate how an complex. A prominent example is the analysis of conditional
input translates into an output via hidden layers of neurons mediation models where mediators and moderators interact.
(also referred to as forward propagation). Using training For example, the indirect effect between an antecedent and a
data where the input and outputs are known, researchers target construct may depend on a moderator, implying that
then estimate the weights that characterize the relationships the mediating effect is conditional upon the values of the
between the nodes. Prediction errors are then used to adjust moderator (Cheah et al. 2021). PLS-SEM proves particularly
the weights ex post (also referred to as backward propa- useful for analyzing such effects. Unlike PROCESS-based
gation); nodes with higher errors are given lower weights, analyses that use the construct scores from a previous analy-
while nodes with smaller errors are weighted more strongly sis as input (Hayes 2018), PLS-SEM estimates reflect the
in the network. Forward and backward propagation alter- entire model structure across different layers of constructs
nate until the model’s prediction error reaches a predeter- while accounting for measurement error (Hair et al. 2022).
mined limit or a set number of iterations is being reached Researchers analyzing mediating effects between latent
(e.g., Hair et al. 2020). The logic behind ANN can readily variables should therefore rely on PLS-SEM (Sarstedt et al.
be transferred to PLS path models where constructs can be 2020). Chang et al.’s (2023) paper published in this special
conceived as neurons linked via inner model weights (i.e., issue considers such a complex mediation model type. Their
path relationships). While ANN allow for the specification analysis shows that the mediating role of engagement in the
of multiple layers, all the neurons from one layer are related relationship between consumers’ technology experience and
to the neurons from the immediate previous and following their intention to use a new technology depends on privacy
Advanced marketing analytics using partial least squares structural equation modeling… 3
concerns. Sarstedt and Moisescu (2023) note that the analy- Analytics Editorial Review Board and therefore served as
sis of such mediating effects can come with a substantial ad hoc reviewers. Thank you for your support! Finally, we
level of uncertainty, especially when more than one con- would like to thank the journal’s editors, Anjala Krishen
figuration of a mediating effect is plausible. To identify the and Maria Petrescu, for inviting us to edit this special
“best” mediation model in such a setting, researchers rely issue. Your support and visionary development of the
on information-theoretic model selection criteria (Burnham PLS-SEM method has moved the field considerably for-
and Anderson 2002), whose values may, however, give false ward. Thank you!
confidence in the results’ adequacy. Addressing this concern,
Funding Open Access funding enabled and organized by Projekt
Sarstedt and Moisescu’s (2023) paper published in this spe- DEAL.
cial issue introduces a procedure to quantify the metrological
uncertainty (Rigdon et al. 2020; Rigdon and Sarstedt 2022) Open Access This article is licensed under a Creative Commons Attri-
inherent in the comparison of different mediation models. bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
Specifically, their procedure uses information-theoretic as you give appropriate credit to the original author(s) and the source,
model selection criteria to weigh model-specific bootstrap provide a link to the Creative Commons licence, and indicate if changes
samples in order to adjust the model parameters’ confidence were made. The images or other third party material in this article are
intervals (Rigdon et al. 2023). included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
The special issue concludes with two additional review the article’s Creative Commons licence and your intended use is not
articles, whose topics are likely to be relevant for many permitted by statutory regulation or exceeds the permitted use, you will
researchers working with PLS-SEM. Cheah et al. (2023) need to obtain permission directly from the copyright holder. To view a
present a review of version 4 of SmartPLS (Ringle et al. copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
2022), which—according to recent reviews (Sarstedt et al.
2022; Wang et al. 2023; Zeng et al. 2021)—is the most fre-
quently used software program for conducting PLS-SEM
analyses. The authors review the program’s core features References
vis-à- vis version 3 (Ringle et al. 2015)—see Sarstedt and
Basco, R., J.F. Hair, C.M. Ringle, and M. Sarstedt. 2022. Advancing
Cheah (2019) for an earlier review. Finally, Gironda (2023) family business research through modeling nonlinear relation-
presents a review of the second edition of Hair et al.’s (2024) ships: Comparing PLS-SEM and multiple regression. Journal of
book Advanced Issues in Partial Least Squares Structural Family Business Strategy 12 (3): 100457.
Equation Modeling (PLS-SEM), which covers several of the Becker, J.-M., D. Proksch, and C.M. Ringle. 2022. Revisiting Gaussian
copulas to handle endogenous regressors. Journal of the Academy
extensions used in this special issue’s papers. of Marketing Science 50 (1): 46–66.
We believe that the research papers of this issue not only Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Multi-
showcase the potentials of using advanced modeling and model Inference, 2nd ed. New York, NY: Springer.
assessment routines in PLS-SEM, but also offer various Chang, Y.-S., R. Konar, J.H. Cheah, and X.-J. Lim. 2023. Does privacy
still matter in smart technology experience? A conditional media-
avenues for further research in this rapidly emerging field. tion analysis. Journal of Marketing Analytics. https://doi.org/10.
For example, some researchers equate formative measure- 1057/s41270-023-00240-8.
ment with composite models—as assumed by PLS-SEM— Cheah, J.H., F. Magno, and F. Cassia. 2023. Reviewing the SmartPLS
and reflective measurement with factor models—as assumed 4 software: The latest features and enhancements. Journal of Mar-
keting Analytics. https://doi.org/10.1057/s41270-023-00266-y.
by covariance structure analysis (Cho et al. 2022) which, Cheah, J.H., C. Nitzl, J.L. Roldán, G. Cepeda-Carrión, and S.P. Gud-
however, confuses the measurement theory layer with the ergan. 2021. A primer on the conditional mediation analysis in
model layer (e.g., Sarstedt et al. 2016). Addressing this con- PLS-SEM. ACM SIGMIS Database: the DATABASE for Advances
cern, more research is needed to conceptually distinguish in Information Systems 52: 43–100.
Cho, G., M. Sarstedt, and H. Hwang. 2022. A comparison of covari-
composite models and factor models. We also believe that ance structure analysis, partial least squares path modeling and
further development of the necessity logic in the context generalized structured component analysis in factor- and com-
of PLS-SEM hold considerable promise, for example, by posite models. British Journal of Mathematical and Statistical
adopting the multiple NCA (Dul 2023). Finally, recent Psychology 75 (2): 220–251.
Crespo, C.F., A.G. Ferreira, and R.M. Cardoso. 2023. The influence
advances in model assessment such as Liengaard et al.’s of storytelling on the consumer–brand relationship experience.
(2021) CVPAT may be extended to offer a broader range of Journal of Marketing Analytics 11 (1): 41–56.
predictive benchmarks for PLS path models. Damberg, S., Y. Lium, and C.M. Ringle. 2023. Does culture matter?
We are grateful to the reviewers who contributed their Corporate reputation and sustainable satisfaction in the Chinese
and German banking sector. Journal of Marketing Analytics.
valuable time and talent to develop this special issue, and https://doi.org/10.1057/s41270-023-00259-x.
ensured the articles’ quality with their constructive com- Damberg, S., M. Schwaiger, and C.M. Ringle. 2022. What’s important
ments and suggestions to the authors. Many of the review- for relationship management? The mediating roles of relational
ers were not regular members of the Journal of Marketing
4 M. Sarstedt, Y. Liu
trust and satisfaction for loyalty of cooperative banks’ customers. PLS-SEM. In H. Latan, J. F. Hair, and R. Noonan (Eds.), Partial
Journal of Marketing Analytics 10 (1): 3–18. Least Squares Path Modeling: Basic Concepts, Methodological
Dul, J. 2016a. Identifying single necessary conditions with NCA and Issues, and Applications (2nd ed.). Cham: Springer.
fsQCA. Journal of Business Research 69 (4): 1516–1523. Richter, N.F., S. Schubring, S. Hauff, C.M. Ringle, and M. Sarstedt.
Dul, J. 2016b. Necessary condition analysis (NCA): Logic and method- 2020. When predictors of outcomes are necessary: Guidelines
ology of “necessary but not sufficient” causality. Organizational for the combined use of PLS-SEM and NCA. Industrial Man-
Research Methods 19 (1): 10–52. agement & Data Systems 120 (12): 2243–2267.
Dul, J. 2023. Necessary condition analysis (NCA) with R (Version Rigdon, E.E., and M. Sarstedt, M. 2022. Accounting for uncertainty
3.3.3): A quick start guide. Available at SSRN: https://ssrn.com/ in the measurement of unobservable marketing phenomena. In:
abstract=2624981 H. Baumgartner and B. Weijters (Eds.), Review of Marketing
Dul, J., S. Hauff, and R.B. Bouncken. 2023. Necessary condition analy- Research, Vol. 19 (pp. 53–73). Bingley, UK: Emerald.
sis (NCA): Review of research topics and guidelines for good Rigdon, E.E., M. Sarstedt, and J.-M. Becker. 2020. Quantify uncer-
practice. Review of Managerial Science 17 (2): 683–714. tainty in behavioral research. Nature Human Behaviour 4:
Gironda, J. 2023. Review of advanced issues in partial least squares 329–331.
structural equation modeling, Second Edition, Journal of Market- Rigdon, E.E., M. Sarstedt, and O.I. Moisescu. 2023. Quantifying model
ing Analytics, Advance online publication. selection uncertainty via bootstrapping and Akaike weights. Inter-
Guenther, P., M. Guenther, C.M. Ringle, G. Zaefarian, and S. Cart- national Journal of Consumer Studies 47 (4): 1596–1608.
wright. 2023. Improving PLS-SEM use for business marketing Ringle, C.M., M. Sarstedt, N. Sinkovics, and R.R. Sinkovics. 2023.
research. Industrial Marketing Management 111: 127–142. A perspective on using partial least squares structural equation
Hair, J.F., and D.E. Harrison. 2021. Essentials of Marketing Analyt- modelling in data articles. Data in Brief 48: 109074.
ics. New York, NY: McGraw-Hill. Ringle, C.M., S. Wende, and J.-M. Becker .2015. SmartPLS 3 [Com-
Hair, J.F., C.L. Hollingsworth, A.B. Randolph, and A.Y.L. Chong. puter software]. Retrieved from http://www.smartpls.com
2017. An updated and expanded assessment of PLS-SEM in Ringle, C.M., S. Wende, and J.-M. Becker. 2022. SmartPLS 4 [Com-
information systems research. Industrial Management & Data puter software]. Retrieved from http://www.smartpls.com
Systems 117 (3): 442–458. Russo, D., and K.-J. Stol. 2021. PLS-SEM for software engineering
Hair, J.F., G.T.M. Hult, C.M. Ringle, and M. Sarstedt. 2022. A research: An introduction and survey. ACM Computing Surveys
primer on partial least squares structural equation modeling 54 (4): 1–38.
(PLS-SEM), 3rd ed. Thousand Oaks, CA: Sage. Sarstedt, M., and J.-H. Cheah. 2019. Partial least squares structural
Hair, J.F., M. Sarstedt, C.M. Ringle, and S.P. Gudergan. 2024. equation modeling using SmartPLS: A software review. Journal
Advanced Issues in Partial Least Squares Structural Equation of Marketing Analytics 7 (3): 196–202.
Modeling (PLS-SEM), 2nd ed. Thousand Oaks, CA: Sage. Sarstedt, M., J.F. Hair, J.-H. Cheah, J.-M. Becker, and C.M. Ringle.
Hayes, A.F. 2018. Introduction to Mediation, Moderation, and Con- 2019. How to specify, estimate, and validate higher-order models.
ditional Process Analysis: A Regression-based Approach. New Australasian Marketing Journal 27 (3): 197–211.
York, NY: The Guilford Press. Sarstedt, M., J.F. Hair, C. Nitzl, C.M. Ringle, and M.C. Howard. 2020.
Leong, K.Y., J.S.Y. Ho, S. Tehseen, E. Yafi, and T.-H. Cham. 2023. Beyond a tandem analysis of SEM and PROCESS: Use PLS-SEM
The intangible values of live streaming and their effect on audi- for mediation analyses! International Journal of Market Research
ence engagement. Journal of Marketing Analytics. https://doi. 62 (3): 288–299.
org/10.1057/s41270-023-00247-1. Sarstedt, M., J.F. Hair, M. Pick, B.D. Liengaard, L. Radomir, and C.M.
Liengaard, B., P.N. Sharma, G.T.M. Hult, M.B. Jensen, M. Sarstedt, Ringle. 2022. Progress in partial least squares structural equation
J.F. Hair, and C.M. Ringle. 2021. Prediction: Coveted, yet for- modeling use in marketing research in the last decade. Psychology
saken? Introducing a cross-validated predictive ability test in & Marketing 39 (5): 1035–1064.
partial least squares path modeling. Decision Sciences 52 (2): Sarstedt, M., J.F. Hair, C.M. Ringle, K.O. Thiele, and S.P. Gudergan.
362–392. 2016. Estimation issues with PLS and CBSEM: Where the bias
Lohmöller, J.-B. 1989. Latent Variable Path Modeling with Partial lies! Journal of Business Research 69 (10): 3998–4010.
Least Squares. Heidelberg: Physica. Sarstedt, M., and O.I. Moisescu. 2023. Quantifying uncertainty in PLS-
Mkedder, N., and F.Z. Özata. 2023. I will buy virtual goods if I SEM-based mediation analyses. Journal of Marketing Analytics.
like them: A hybrid PLS-SEM-artificial neural network (ANN) https://doi.org/10.1057/s41270-023-00231-9.
analytical approach. Journal of Marketing Analytics. https://doi. Sharma, P.N., B.D. Liengaard, J.F. Hair, M. Sarstedt, and C.M. Ringle.
org/10.1057/s41270-023-00252-4. 2023. Predictive model assessment and selection in composite-
Nitzl, C., J.L. Roldan, and G. Cepeda. 2016. Mediation analysis in based modeling using PLS-SEM: Extensions and guidelines for
partial least squares path modeling: Helping researchers discuss using CVPAT. European Journal of Marketing 57 (6): 1662–1677.
more sophisticated models. Industrial Management & Data Sys- Shmueli, G., S. Ray, J.M. Velasquez Estrada, and S.B. Chatla. 2016.
tems 116 (9): 1849–1864. The elephant in the room: Evaluating the predictive performance
Petter, S. 2018. “Haters gonna hate”: PLS and information sys- of PLS models. Journal of Business Research 69 (10): 4552–4564.
tems research. ACM SIGMIS Database: THe DATABASE for Shmueli, G., M. Sarstedt, J.F. Hair, J.-H. Cheah, H. Ting, and C.M.
Advances in Information Systems 49 (2): 10–13. Ringle. 2019. Predictive model assessment in PLS-SEM: Guide-
Petter, S., Y. Hadavi, and Y. 2021. With great power comes great lines for using PLSpredict. European Journal of Marketing 53
responsibility: The use of partial least squares in information (11): 2322–2347.
systems research. ACM SIGMIS Database: the DATABASE for Soares, J.C., R. Limongi, J.H. De Sousa Júnior, W. Soares Santos,
Advances in Information Systems 52: 10–23. M. Raasch, and L. Hoeckesfeld. 2023. Assessing the effects of
Richter, N.F., S. Hauff, C.M. Ringle, and S.P. Gudergan. 2022. The COVID-19-related risk on online shopping behavior. Journal of
use of partial least squares structural equation modeling and Marketing Analytics 11 (1): 82–94.
complementary methods in international management research. Sternad Zabukovšek, S., S. Bobek, U. Zabukovšek, Z. Kalinić, and P.
Management International Review 62 (4): 449–470. Tominc. 2022. Enhancing PLS-SEM-enabled research with ANN
Richter, N.F., S. Hauff, C.M. Ringle, M. Sarstedt, A. Kolev, and S. and IPMA: Research study of enterprise resource planning (ERP)
Schubring. 2023. How to apply necessary condition analysis in
Advanced marketing analytics using partial least squares structural equation modeling… 5
systems’ acceptance based on the technology acceptance model main research interest is the advancement of research methods to fur-
(TAM). Mathematics 10 (9): 1379. ther the understanding of consumer behavior. His research has been
Tiwari, P., R.P.S. Kaurav, and K.Y. Koay. 2023. Understand- published in Nature Human Behavior, Journal of Marketing Research,
ing travel apps usage intention: Findings from PLS and NCA. Journal of the Academy of Marketing Science, Multivariate Behavioral
Journal of Marketing Analytics. https:// d oi. o rg/ 1 0. 1 057/ Research, Organizational Research Methods, MIS Quarterly, and Psy-
s41270-023-00258-y. chometrika, among others. Marko has been repeatedly named member
Wang, S., J.-H. Cheah, C.Y. Wong, and T. Ramayah. 2023. Pro- of Clarivate Analytics’ Highly Cited Researchers List, which includes
gress in partial least squares structural equation modeling use the “world’s most impactful scientific researchers.” In March 2022,
in logistics and supply chain management in the last decade: A he was awarded an honorary doctorate from Babeș-Bolyai-University
structured literature review. International Journal of Physical Cluj-Napoca for his research achievements and contributions to inter-
Distribution & Logistics Management. https://doi.org/10.1108/ national exchange.
IJPDLM-06-2023-0200.
Wold, H. 1982. Soft modeling: The basic design and some extensions. Yide Liu is a Professor at Macau University of Science and Technol-
In K.G. Jöreskog and H. Wold (Eds.), Systems Under Indirect ogy. He focuses on innovation management research. His research was
Observations. Part II (pp. 1–54). Amsterdam: North-Holland. published in academic journals including Journal of Business Research,
Zeng, N., Y. Liu, P. Gong, M. Hertogh, and M. König. 2021. Do right International Journal of Contemporary Hospitality Management, Inter-
PLS and do PLS right: A critical review of the application of net Research, etc. Clarivate Analytics has listed two of his research as
PLS-SEM in construction management research. Frontiers of highly cited and hot papers. Yide Liu co-chaired a series of interna-
Engineering Management 8: 356–369. tional conferences on innovation, information systems and quantitative
methods. Yide Liu provided technical consulting services to multiple
Publisher's Note Springer Nature remains neutral with regard to firms. He has served as the Chief Operating Officer for one emerging
jurisdictional claims in published maps and institutional affiliations. growth technology company.
Marko Sarstedt is a Chaired Professor of marketing at the Ludwig-
Maximilians-University Munich (Germany) and an adjunct research
professor at Babeș-Bolyai-University Cluj-Napoca (Romania). His
