Investigating the reliable and interpretable components existing among twenty-one variables using principal component analysis

The study investgated the reliable and interpretable components existing among twenty-one variables. This study used principal components analysis to obtain uncorrelated linear combinations of the original variables that account for as much of the total variance in the original variables as possible. Three hundred and fifty-four (354) elementary preservice mathematics teachers from five colleges (college 1 = 35 males and 35 females; college 2 = 35 males and 36 females; college 3= 38 males and 35 females; college 4 = 37 males and 38 females; and college 5= 33 males and 32 females) of education in Ghana participated in the study. The study identified eight (8) components or factors (exertion of authority, setting example, likeability, teacher posture, teacher attitude, student concern, teacher expectation, and setting the ground rules) as against three (3) components or factors (R = rule-based; D = dominance; N = nurturance) in the Evertson et al.’s (1989) study. The difference in interpretability of the components could stem from the cultural, geographical, and even economical settings that contributed for participants to respond in a particular way. This study demonstrates that even though validated questionnaire or instrument is recommended for empirical research, care should be taken when conducting research with the same questionnaire or instrument in another geographical location.