Endnotes

---

¹ It is important to note in this regard that continued use (Bhattacherjee, 2001) is not the same as habit. Habit is what one usually does; it is a behavioral preference in the present. It is not an intended behavior in the future.

² Construct Reliability is calculated as: (Σ std loadings)² / ((Σ std loadings)² + Σ (std errors))

³ It should be noted that almost all TAM studies applying LISREL used only partial scales of PU and PEOU. Segars & Grover (1993) used a five item scale of PU and a three item scale of PEOU; Subramanian (1994) used a three item scale of PU and a four item scale of PEOU; Gefen & Keil (1998)) used a five item PU scale and a three item PEOU scale; and Gefen (2000) used a four items scale of PU and a two item scale of PEOU.

---