My PhD Journey

When performing statistical analysis using structural equation modelling, specifically in PLSSEM, there are several approaches for moderation analysis, including: Product- Indicator Two Stage Orthogonalizing  Prior to that, researcher must identify his or her theoretical frame falls under reflective construct or  just formative construct. Depending on researcher’s objective, he or she can choose any of the three approaches. 2-stage approaches has highest power, it is the most likely approaches to create significant attraction. It is recommended to make use of 2-stage approaches and make analysis. I f researcher has smallest population, he or she suppose to use orthogonalizing because the interaction effect size is the most correct and it maximises explained variance in his or her dependent variable. Product indicator approach has no particular advantage but still if it is formative construct can use product indicator approach Formative cons...

First researcher should focus on the significance of the moderating effect (Z) To clarify, it is possible that a moderator variable (M) may or may not have an effect on the dependent variable (Y) Thus, the decision as to whether there is any moderating effect should be made based on a significance relationship between the moderating effect (Z) and the dependent variable (Y)     Second , researchers must calculate and report the effect size (f 2), and how much it contributes to R2 as a function of the moderator Only a few software packages (e.g. Smart PLS3.0) calculate f 2 by default For other there are online spreadsheets which can be used to calculate effect size (see: http://statwiki.kolobkreations.com ) Lastly, researchers must execute and report a simple slope plot for visual inspection of the direction and strength of the moderating effect SmartPLS users can check out a simple slope plot under ‘Final Results’ and ‘Simple Slope Analysis...

Theoretically moderator effect differs slightly from interaction effects Moderator refer to variable that alter an observed relation in the population While interaction effect refer to any situation in which the effect of one variable depends on the level of another variable   Moderator is not part of a causal sequence but qualifies the relation between X and Y For intervention research, moderator variables may reflect subgroups of persons for which the intervention is more or less effective than for other groups In general, moderator variables are critical for understanding the generalisability of a research findings to subgroups Example: The true relationship between X & Y is more revealed when critical moderating variables are inserted in the model It is point out that ‘size of the hotel’ is an influential factor which effects the relationship between X & Y, because small hotels are much more exposed to risks than large franchise hotels. (Kang e...

Case 1: Moderator and IV are Categorical A dichotomous independent variable’s effect on the dependent variable varies as a function of another dichotomy 2 x 2 ANOVA The moderation is indicated by the interaction IV – Level of Education DV – Income Case 2: Moderator is Categorical and IV is Continuos Deficiency: Assumes that the IV has equal variance at each level of the moderator The effect of IV on DV is tested using unstandardized regression coefficient. The regression coefficients are then tested for differences (see formula, Cohen & Cohen, 1983, p.56) Reliabilities should be tested foe the level of moderation, and slopes should be disattenuated Case 3: Moderator is continuous and IV is categorical We must know a prior how the effect of the IV varies as a function of the moderator (1)   Linear Function (2)   Step Function (3)   Quadratic Function Example: IV: Rational VAS Fear – arousing attitud...