Content area
Abstract
Procedures. This study established the maximized coefficient alpha for the Principles of Adult Learning Scale (PALS), its seven factors, and the 44 individual items using Pepin's formula. Additionally, the factor comparability of the scale was determined using LISREL for data-reduction and hypothesis testing. The subjects for the study were 593 adult education instructors and 809 higher education instructors.
Results. The maximized alpha level for the PALS was 0.97 for the adult education instructor group and 0.96 for the higher education instructor group. Satisfactory maximized alpha values were reached for: Factor 1 (Learner-Centered) 0.97 (adult education instructor group) and 0.96 (higher education instructor group); Factor 2 (Personalizing Instruction) 0.93 (adult group) and 0.94 (higher education group); and Factor 3 (Relating to Experience) 0.81 (higher education group)
Maximized alpha levels for each of the 44 items reached the required 0.80 level for both groups. Alpha values ranged from 0.80 to 0.98.
Comparing factor scores of the two groups yielded significant differences for Factor 1 $(p <$.0001), Factor 4, Assessing Student Needs, ($p <$.01), Factor 5, Climate Building, ($p <$.01), and Factor 3 ($p <$.05).
An analysis of variance was conducted on the total PALS scores for the two groups. The procedure yielded a significant difference between the two respondent groups (df = 1387, F = 28.54, $p <$.0001)
Conclusions. The PALS is a reliable instrument with a maximized alpha level of 0.96. Only three of the seven factors achieved satisfactory alpha levels in at least one of the respondent groups. The PALS is not population invariant. The adult instructor group ($M$ = 144, $SD$ = 17.4) and the higher education instructor group ($M$ = 129, $SD$ = 21.6) were significantly different ($p <$.0001).
