latent profile analysis
Definition
Latent profile analysis is a person-centered statistical method that identifies distinct subgroups within a population based on patterns of responses across multiple continuous or categorical indicators. Rather than treating individual variables in isolation, the approach recovers a discrete latent structure by grouping individuals whose response profiles are similar to one another and dissimilar to those of other groups. Applied to a representative quota sample of 506 non-Hispanic White Americans across the 2024 U.S. presidential election cycle, the method revealed a four-profile solution capturing variation in subjective racial economic status, including a distinct 'last place (tied)' group comprising approximately 15% of the sample who perceived themselves as falling behind both White and Asian Americans while tied with Black Americans.
Sources: Kukharkin et al. (2026)
Related Terms
- political psychology (1 shared article)
- status (1 shared article)
- economic inequality (1 shared article)
- longitudinal design (1 shared article)
Applications
Latent Profile Analysis and Subjective Social Status
Latent profile analysis identified four discrete profiles of subjective status among White Americans based on their perceived position within the racial economic hierarchy relative to their own group and to Black, Hispanic, and Asian Americans. The 'last place (tied)' profile captured individuals who felt simultaneously worse off than stereotypically high-status groups and at risk of being passed by racially marginalized groups.
Sources: Kukharkin et al. (2026)
Latent Profile Analysis and Political Attitudes
Profile membership derived from latent profile analysis predicted meaningful differences in political attitudes and behavior. White Americans with the highest probability of belonging to the 'last place (tied)' profile showed the greatest support for alt-right ideology, DEI bans, and Donald Trump, and were significantly more likely to vote for Trump in the 2024 election than members of the 'third place-dispersed' or 'second place (tied)' profiles, with these associations remaining temporally stable across five waves from early September through election day.
Sources: Kukharkin et al. (2026)
