Colloquium The role of attitudes in the TAM: A theoretically unnecessary construct? Luis Miguel López-Bonilla and Jesús Manuel López-Bonilla Address for correspondence: Dr Luis Miguel López-Bonilla, Departamento de Administración de Empresas y Marketing, Facultad de Ciencias Económicas y Empresariales, Universidad de Sevilla, Avenida Ramón y Cajal, n°1-41018 Seville, Spain. Email: luismi@us.es Introduction As pointed out by Schepers and Wetzel (2007), many studies focus on Technology Acceptance Model (TAM), but some aspects still remain rather unclear. One of these aspects is the role played by attitudes in this model. In two recent contributions to the British Journal of Educational Tech- nology colloquium, Teo (2009) and Nistor and Heymann (2010) find that attitude is unnecessary for the TAM. The main objective of this research is to test the inclusion of attitudes in TAM. Two versions of this model are compared: with attitudes or original model (TAM-O) and without attitudes or revised model (TAM-R). TAM with and without attitudes are applied to the adoption of the student enrolment through three different technologies (Internet, electronic kiosk and mobile phone). These three settings are voluntary, that is, the students could choose between electronic and traditional enrolment. Method Participants A sample of 1414 undergraduate students from a big south European university (University of Seville, Spain) was obtained for this study.This sample is spread over 26 centers, grouped into four main fields (Economics/Law, Humanities, Sciences/Technical Studies and Biology/Medicine).The sample elements are extracted in proportion to the relative size of each area. Participants were 488 students in the first setting or e-service (electronic kiosk) with 56.97% of females and 43.03% of males, 456 students in the second e-service (personal computer), with 57.89% of females and 42.11% of males and 460 students in the third e-service (mobile telephone), with 56.3% of females and 43.7% of males. Instruments Constructs are measured on 12 items adapted from published studies (eg, Dabholkar, 1994, 1996; Davis, 1989; Davis, Bagozzi & Warshaw, 1989; Fishbein & Ajzen, 1975). All items were measured on a 7-point semantic differential scale of three items. SmartPLS (Hamburg, Germany) package is used to analyze the results of the questionnaires. SmartPLS is developed by Ringle, Wende and Will (2005). Analytical procedure A limited estimation method is used, partial least squares (PLS), denominated in this way by dividing the list of parameters to be estimated into various subsets by means of using multiple linear regressions that employ the algorithm of the ordinary least squares. There is no overall fit index in PLS path modeling. Nevertheless, a global criterion of goodness of fit (GoF) has been proposed by Amato, Vinzi and Tenenhaus (2005). Such index has been devel- oped in order to take into account the model performance in both the measurement and the structural model. For this reason, the GoF index is obtained as the geometric mean of the average British Journal of Educational Technology Vol 42 No 6 2011 E160–E162 doi:10.1111/j.1467-8535.2011.01232.x © 2011 The Authors. British Journal of Educational Technology © 2011 BERA. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.