Abstract
The problems for statistical analysis posed by the groups vs individuals design are presented, a statistical model for the design is developed, and 2 alternative solutions are proposed--the pseudogroup and quasi-F procedures. Solutions are extended to the general case of designs including several levels of the group-size variable. The quasi-F approach is applied to the problem of contrasting combinations of means from several group-size conditions. (14 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).
Original language | English |
---|---|
Pages (from-to) | 86-98 |
Number of pages | 13 |
Journal | Journal of Personality and Social Psychology |
Volume | 40 |
Issue number | 1 |
DOIs | |
State | Published - Jan 1981 |
Keywords
- pseudogroup & quasi-F statistical analyses of group vs individuals design
ASJC Scopus subject areas
- Social Psychology
- Sociology and Political Science