This research examined several independent variables and their prediction of perceived freedom in leisure (PFL). Four instruments were utilized to collect data from research subjects regarding the independent variables of shyness, extroversion, gender and activity preference style and the dependent variable, PFL. Cronbach's alpha reliability coefficients were calculated for each scale employed in the research. Reliabilities for the scales within this research were as follows: Stanford Shyness Survey (.78), Adult Short Form of the Leisure Diagnostic Battery (.92), three scales from the Eysenck Personality Questionnaire - Extroversion (.85), Neuroticism (.79), and LIE (.75), and the Activity Preference Style Scales - Active (.45), Group (.53), and Risk (.67). Due to the low alpha reliabilities of two of the Activity Preference Style Scales, Active and Group, factor analysis was performed in an attempt to construct new sub-scales with higher alpha reliabilities. This resulted in some of the new sub-scales, as well as the original Active and Group scales being used in the data analysis. The sample was comprised of 325 undergraduate students enrolled in a required history or English class. The age of the sample ranged from 17 to 50 with a mean age of 20.4. Questionnaires were given out during class time and students were instructed to complete them at home and return them to their instructor. Respondents were categorized into six non-independent groups: all subjects, not shy subjects, shy subjects, and three shyness sub-groups — shy now and in the past, shy now but not in the past and shy in the past. Stepwise multiple linear regression analysis was employed with four different sets of Activity Preference Style scales and sub-scales in the prediction of PFL for the six subject groupings. Regardless of a subject's level of shyness, extroversion, and in several other cases, one of the activity style variables were the only significant predictors of PFL. The best predictor of PFL for not shy subjects was risk.