Psychometric analysis of the Food Technology Neophobia Scale in a Chilean sample

- The psychometric properties of the Food Technology Neophobia Scale were assessed in a developing South American country.
- The original FTNS model was tested, along with seven subsequent models.
- A confirmatory factor analysis testing the factorial invariance by student/working adult status showed a good fit to data.
- An abbreviated version of the FTNS is proposed.

The aim of this study was to determine the psychometric properties of the Food Technology Neophobia Scale (FTNS) in a developing South American country. A convenience sample of 400 adults from Temuco, Chile, was obtained. Two hundred subjects were surveyed at a supermarket and 200 were university students. After a data cleansing process, a sample of 332 subjects was obtained. The questionnaire included the 13-item FTNS and socio-demographic questions. Using confirmatory factor analysis (CFA), the original FTNS model proposed by Cox and Evans (2008) was tested, along with other seven models that resulted from statistically based variations. There was no evidence of validity for the original model, and the remaining models also showed a bad fit to data. However, a CFA was conducted testing the factorial invariance by student or working adult status. Thus, a one-dimensional model was tested, which showed a good fit to data and presented a good reliability for students and non-students. This model presented strong factorial invariance. To encourage parsimony, this model is proposed as an abbreviated version of the FTNS (AFTNS), which is reduced from 13 to 9 items and from four to one factor. In order to obtain improved goodness of fit indexes, it is recommended to consider the categorical variable student/non-student in the analysis if this variable is present in the data, especially for the use of the AFTNS in populations with similar characteristics.