Outer Loadings : Original Sample, Sample Mean, Stdv..

[binosjourney]

Hello,

I have several questions as this is my first time using this software.
- As it stated in the bootstrapping result that T-statics is derived from dividing the original sample with the standard deviation. However when I try to manually calculate it, it gave me slightly different result of T-statistics from what the software has calculated e.g : BEI1<--BEI Tstats = 43.2857143 (manually calculated) and 43.695 (provided by SmartPLS) some even has greater diference. You can take a look at the data with the provided image below




- What is sample mean? is it derived from the bootstrapping re sample or is it provided from the original sample?

I need to convince my thesis supervisor with this two because he ask this last time I meet him, thank you.

[jmbecker]

The difference should result from rounding errors. You look at results with a precision of 3 decimal points. Internally we are using double (i.e., which precision is roughly at 16 decimal points).

The mean estimate is the mean of the bootstrap distribution. This can be slightly different from the original sample estimate. The difference is called bias and is, beside other things, used to calculate the bias-corrected confidence intervals.