Mediated relationships with TV characters

Five TV actors completed the Big Five personality scale for a character they played on a popular Israeli TV comedy. Viewers of each of these series completed the same scales both for themselves and as they thought the characters would have completed them. They then completed parasocial relationship and identification scales with respect to the same character. Perceived and measured similarity scores (i.e., using the actors’ scores) were computed for each viewer-character pair. These similarity scores were then used to predict both parasocial relationship strength and the degree of identification. Results show that perceived and measured similarity are mostly unrelated and that perceived similarity, but not measured similarity, is related to parasocial relationships and identification. Implications of these results for mediated relationships theory and measurement validity are discussed.

Subnormal Inductive Sources and π-Partial Characters

A character pair (H,θ) in a group G consists of a subgroup H and a character θ ∈ Irr (H). A character pair (H,θ) is an inductive source in G if induction to G defines an injective map from the irreducible characters of T, the stabilizer of (H,θ), that lie over θ into Irr (G). Let π be a set of primes, and suppose G is π-separable. We consider Isaacs' π-partial characters and their canonical lifts. If θ ∈ Irr (H) is such a lift, then the restriction θ0 of θ to the π-elements of H is an irreducible π-partial character of H. In this paper, when (H,θ) is an inductive source so that H is subnormal and θ is a canonical lift, we show that induction is an injection from the irreducible π-partial characters of T that lie over θ0 to the irreducible π-partial characters of G. We apply this to obtain a partial generalization of Isaacs' nucleus of a character, and present several examples to see what can go wrong with our generalization.