Presentation is loading. Please wait.

Presentation is loading. Please wait.

Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program PLS-SEM: Introduction Continued (Part 2)

Similar presentations


Presentation on theme: "Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program PLS-SEM: Introduction Continued (Part 2)"— Presentation transcript:

1 Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program PLS-SEM: Introduction Continued (Part 2)

2 Systematic Process for applying PLS-SEM

3 Significance of PLS-SEM Parameters = Bootstrapping PLS-SEM does not assume the data is normally distributed, which implies that parametric significance tests used in regression analyses cannot be applied to test whether coefficients such as outer weights and loadings are significant. Instead, PLS-SEM relies on a nonparametric bootstrap procedure to test coefficients for their significance. In bootstrapping, a large number of subsamples (i.e., bootstrap samples) is drawn from the original sample with replacement. Replacement means that each time an observation is drawn at random from the sampling population, it is returned to the sampling population before the next observation is drawn (i.e., the population from which the observations are drawn always contains all the same elements). Therefore, an observation for a certain subsample can be selected more than once, or may not be selected at all for another subsample. The number of bootstrap samples should be high but must be at least equal to the number of valid observations in the dataset. The recommended number of bootstrap samples is 5,000.

4 SmartPLS Bootstrapping Bootstrapping estimates a PLS path model for each subsample: Bootstrapping estimates a PLS path model for each subsample: Samples: Number of random samples drawn from the original sample (at minimum should equal the number of observations in the original sample, but 5,000 is recommended). Samples: Number of random samples drawn from the original sample (at minimum should equal the number of observations in the original sample, but 5,000 is recommended). Cases: Number of cases drawn in each sample run (should be at least as large as the number of valid observations in the original sample). Cases: Number of cases drawn in each sample run (should be at least as large as the number of valid observations in the original sample). Bootstrapping provides mean values and standard errors for: Bootstrapping provides mean values and standard errors for: inner model path coefficients. inner model path coefficients. weights and loadings in the measurement models. weights and loadings in the measurement models. Use bootstrapping

5 If you have missing data do not use mean replacement because bootstrapping draws samples with replacement. Use Casewise Replacement. If you have missing data do not use mean replacement because bootstrapping draws samples with replacement. Use Casewise Replacement. Use individual (sign) changes option Make sure the number of cases are equal to the number of valid observations in your dataset.Make sure the number of cases are equal to the number of valid observations in your dataset. Set cases = samples size (or higher)Set cases = samples size (or higher) Caution!!! It is a common mistake to set samples equal to the overall number of observations. SmartPLS Bootstrapping

6 Make sure the number of cases are equal to (or more than) the number of valid observations in your dataset. Set cases = sample size (or higher). Note that the number is now 344.Make sure the number of cases are equal to (or more than) the number of valid observations in your dataset. Set cases = sample size (or higher). Note that the number is now 344. We have also set the number of samples as 5,000.We have also set the number of samples as 5,000.

7 Bootstrapping HTML Report – Table of Contents Click on to access HTML report

8 Significant t-values 1.65 for 10%1.65 for 10% 1.96 for 5%1.96 for 5% 2.58 for 1%2.58 for 1% (all two-tailed) Results based on Cases = 344 and Samples = 5,000 Bootstrapping Option (Total Effects tables) – Significance of Structural Path Coefficients

9 Results based on Cases = 344 and Samples = 5,000 Bootstrapping Option – Significance of Indicator Loadings

10 SmartPLS Predictive Relevance – Blindfolding o Q² is a criterion to evaluate how well the model predicts the data of omitted cases. It is referred to as predictive relevance. o The process involves omitting (removing) or blindfolding one case at a time and re-estimating the model parameters based on the remaining cases. The omitted case values are then predicted on the basis of the newly estimated parameters of the remaining cases. o Procedure: Set an omission distance D. Note: The number of cases in your data must not be a multiple integer number of the omission distance (otherwise the blindfolding procedure yields erroneous results). Experience has shown that d values between 5 and 10 typically work well.Set an omission distance D. Note: The number of cases in your data must not be a multiple integer number of the omission distance (otherwise the blindfolding procedure yields erroneous results). Experience has shown that d values between 5 and 10 typically work well. Interpret the cross-validated redundancy, because it uses the PLS-SEM estimates of both the structural model and the measurement models for data prediction. Also, in most instances the focus is on predicting the data of the target endogenous constructs.Interpret the cross-validated redundancy, because it uses the PLS-SEM estimates of both the structural model and the measurement models for data prediction. Also, in most instances the focus is on predicting the data of the target endogenous constructs.

11 11 LV 3 MV 1 MV 2 MV 3 LV 1 LV 2 LV 3 MV 1 MV 2 MV 3 LV 1 LV 2 Step 1: The scores of the endogenous LV(s) are estimated using the scores of the exogenous LVs Step 2: Newly estimated LV scores are used to estimate the missing MV data Cross-validated redundancy Cross-validated communality Only step 2. SmartPLS Predictive Relevance – Blindfolding Redundancy vs. Communality?

12 SmartPLS Results – Blindfolding Use blindfolding

13 SmartPLS Results – Blindfolding Make sure that n / Omission distance is not an integer (here: n = 344). (here: n = 344). Check all boxes

14 Click on to access HTML report Click on to access HTML report SmartPLS Results – Blindfolding

15 Click on Construct Crossvalidated Redundancy Predictive relevance is demonstrated for both endogenous constructs. SmartPLS Results – Blindfolding Q² > 0: model has predictive relevance. Q² 0 or Q² < 0: model is lacking predictive relevance.

16 * Ranking based on Hult et al. (2009) PLS-SEM and Research in Marketing Top 30 marketing journals* – 204 articles / 311 modelsTop 30 marketing journals* – 204 articles / 311 models 80% of articles published since 2000, 35% in JM, IMM & EJM80% of articles published since 2000, 35% in JM, IMM & EJM 2010 = 25% Totals for 5 year periods Individual years An Assessment of the Use of Partial Least Squares Structural Equation Modeling in Marketing Research, JAMS, Vol. 40 (3), May 2012.

17 PLS-SEM and Research in Marketing Reasons for using PLS – non-normal data (50%), small sample size (46%), formative measures (33%), prediction = research objective (28%), complex models (13%), categorical variables (13%).Reasons for using PLS – non-normal data (50%), small sample size (46%), formative measures (33%), prediction = research objective (28%), complex models (13%), categorical variables (13%). Average PLS sample size is 211 compared to 246 for CB-SEM. But 25% had less than 100 observations, and 9% did not meet recommended sample size criteria.Average PLS sample size is 211 compared to 246 for CB-SEM. But 25% had less than 100 observations, and 9% did not meet recommended sample size criteria. No studies report skewness or kurtosis.No studies report skewness or kurtosis. 42% reflective only; 6% formative only; 40% mixed; 12% no indication.42% reflective only; 6% formative only; 40% mixed; 12% no indication.

18 PLS-SEM and Research in Marketing Outer Model Evaluation ReliabilityReliability 70% reported (56% composite reliability) 70% reported (56% composite reliability) 46% included formative constructs, but 23% used reflective criteria on formative constructs 46% included formative constructs, but 23% used reflective criteria on formative constructs Convergent Validity (AVE) – 57%Convergent Validity (AVE) – 57% Discriminant ValidityDiscriminant Validity Fornell-Larcker – 44% Fornell-Larcker – 44% Cross-loadings – 5% Cross-loadings – 5% Both – 12% Both – 12% Significance of formative indicator weights – 17%Significance of formative indicator weights – 17% Inner Model Evaluation R 2 – 88%R 2 – 88% Predictive relevance (Q 2 ) – 16% (for endogenous variables)Predictive relevance (Q 2 ) – 16% (for endogenous variables) Path coefficient sizes – 96%Path coefficient sizes – 96% Path coefficient significance – 92%Path coefficient significance – 92%

19 Observations and Conclusions PLS-SEM = rapidly emerging tool in marketing literature because... PLS-SEM = rapidly emerging tool in marketing literature because... Flexible data distribution and scaling requirements. Flexible data distribution and scaling requirements. Achieves high levels of statistical power with smaller sample sizes and complex models. Achieves high levels of statistical power with smaller sample sizes and complex models. With complex models produces superior results to CB-SEM. With complex models produces superior results to CB-SEM. Easily handles both reflective and formative measured constructs. Easily handles both reflective and formative measured constructs. PLS-SEMs methodological properties are widely misunderstood (CB-SEM bias). PLS-SEMs methodological properties are widely misunderstood (CB-SEM bias). Marketing scholars need to become familiar with advantages and limitations. Marketing scholars need to become familiar with advantages and limitations.

20 Special Issue, PLS in Marketing, March 2011 Special Issue, PLS in Marketing, March 2011 Hair, Joseph F., Christian M. Ringle, and Marko Sarstedt. PLS-SEM: Indeed a Silver Bullet. Hair, Joseph F., Christian M. Ringle, and Marko Sarstedt. PLS-SEM: Indeed a Silver Bullet. Haenlein, Michael and Andreas M. Kaplan. The Influence of Observed Heterogeneity on Path Coefficient Significance: Technology Acceptance within the Marketing Discipline. Haenlein, Michael and Andreas M. Kaplan. The Influence of Observed Heterogeneity on Path Coefficient Significance: Technology Acceptance within the Marketing Discipline. Eggert, Andreas and Murat Serdaroglu. Exploring the Impact of Sales Technology on Salesperson Performance: A Task-Based Approach. Eggert, Andreas and Murat Serdaroglu. Exploring the Impact of Sales Technology on Salesperson Performance: A Task-Based Approach. Navarro, Antonio, Francisco J. Acedo, Fernando Losada, and Emilio Ruzo. Integrated Model of Export Activity: Analysis of Heterogeneity in Managers Orientations and Perceptions on Strategic Marketing Management in Foreign Markets. Navarro, Antonio, Francisco J. Acedo, Fernando Losada, and Emilio Ruzo. Integrated Model of Export Activity: Analysis of Heterogeneity in Managers Orientations and Perceptions on Strategic Marketing Management in Foreign Markets. Wiedmann, Klaus-Peter, Nadine Hennigs, Steffen Schmidt, and Thomas Wuestefeld. Drivers and Outcomes of Brand Heritage: Consumers Perception of Heritage Brands in the Automotive Industry. Wiedmann, Klaus-Peter, Nadine Hennigs, Steffen Schmidt, and Thomas Wuestefeld. Drivers and Outcomes of Brand Heritage: Consumers Perception of Heritage Brands in the Automotive Industry. Anderson, Rolph, and Srinivasan Swaminathan. Customer Satisfaction and Loyalty in e-Markets: A PLS Path Modeling Approach. Anderson, Rolph, and Srinivasan Swaminathan. Customer Satisfaction and Loyalty in e-Markets: A PLS Path Modeling Approach. Hoffmann, Stefan, Robert Mai, and Maria Smirnova. Development and Validation of a Cross-Nationally Stable Scale of Consumer Animosity. Hoffmann, Stefan, Robert Mai, and Maria Smirnova. Development and Validation of a Cross-Nationally Stable Scale of Consumer Animosity.

21 Other Sources: An Assessment of the Use of Partial Least Squares Structural Equation Modeling In Marketing Research, JAMS, Vol 40 (3), May 2012; Special Issue, LRP, forthcoming 2013, PLS in Long Range Planning. Book: A Primer on Partial Least Squares, Sage, forthcoming 2013.

22 Criteria Variance-Based Modeling (e.g. SmartPLS, PLS Graph) Covariance-Based Modeling (e.g. LISREL, AMOS, Mplus) Objective Prediction oriented Prediction oriented Parameter oriented Distribution Assumptions Non-parametric Non-parametric Normal distribution (parametric) Required sample size Small (min. 30 – 100) Small (min. 30 – 100) High (min. 100 – 800) Model complexity Large models OK Large models OK Large models problematic (50+ indicator variables) Parameter Estimates Potential Bias Potential Bias Stable, if assumptions met Indicators per construct One – two OK Large number OK Typically 3 – 4 minimum to meet identification requirements Statistical tests for parameter estimates Inference requires Jackknifing or Bootstrapping Assumptions must be met Measurement Model Formative and Reflective indicators OK Formative and Reflective indicators OK Typically only Reflective indicators Goodness-of-fit measures NoneMany Summary Comparison: PLS-SEM vs. CB-SEM

23 Sample Size Determination – PLS-SEM Sample size should be equal to the larger of: ten times the largest number of formative indicators used to measure a single construct, orten times the largest number of formative indicators used to measure a single construct, or ten times the largest number of structural paths directed at a particular latent construct in the structural model.ten times the largest number of structural paths directed at a particular latent construct in the structural model.

24 Sample Size Guidelines – PLS-SEM The overall complexity of a structural model has little influence on the sample size requirements for PLS-SEM. The reason is the algorithm does not compute all relationships in the structural model at the same time. Instead, it uses OLS to estimate the SEM models partial regression relationships. Two early studies systematically evaluated the performance of PLS-SEM with small sample sizes and concluded it performed well (e.g., Chin & Newsted, 1999; Hui & Wold, 1982). More recently a simulation study by Reinartz et al. (2009) indicated that PLS-SEM is a good choice when the sample size is small. Moreover, compared to its covariance-based counterpart, PLS-SEM has higher levels of statistical power in situations with complex model structures or smaller sample sizes.

25 HBAT Path Model and Data for PLS-SEM Hypothetical Example


Download ppt "Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program Joe F. Hair, Jr. Founder & Senior Scholar, DBA Program PLS-SEM: Introduction Continued (Part 2)"

Similar presentations


Ads by Google