This transversal study is framed within Sociolinguistics given the study’s subject: Covert prestige in language attitudes, focused on Sociolinguistics’ methodologies to improve our understanding of this phenomenon, providing a prospective nature. It’s based on quantitative and qualitative data’s analysis obtained from a large online anonymous survey in Murcia, a very stigmatized province from Spain (Muñoz Valero, 2019). This survey counts with 661 respondents and several sociodemographic variables were analysed: Sex, age, qualification, background, parents’ background and social network.

Respondent’s behaviour against their vernacular variety despite their maintenance corroborates covert prestige, a phenomenon detected before here (Boluda Nicolás, 2004; Bañón Hernández, 1993; Hernández Campoy & Jiménez Cano, 2004a, 2004b). According to Trafimov, Sheeran, Lombardo, Finaly, Brown & Armitage (2004), language attitudes are divided into two types depending on their nature: (1) Affective components, based on personal experiences, and (2) cognoscitive components, connected to language models which also connect to sociolinguistic awareness (López Morales, 2004). When language models differ much from the vernacular variety, attitudes may work disharmoniously due to high levels of insecurity (Baker, 1992), causing covert prestige (Chambers, 1995).

This research's results provided three detected language-prestige’s types depending on components’ combination: (1) Both positive affective and cognoscitive component lead to positive open language prestige, consequently high language security levels, loyalty and maintenance; (2) Both negative affective component and cognoscitive component lead to high insecurity levels and a likely trend to standardization and dialectal/language death; (3) Positive affective components + negative cognoscitive components = covert prestige and, depending on the case, causing either  language loyalty with high levels of insecurity, partial/moderate standardization or bidialectalism functioning as diglossia. Thus, my conclusions show how any language attitude research related to language insecurity must consider these components regarding methodology’s design and analysis for a better understanding of each case and its implications.