THE EQUAL-INTERVAL NATURE OF SEMANTIC DIFFERENTIAL SCALES - AN EMPIRICAL-INVESTIGATION USING FIELDERS LEAST PREFERRED COWORKER (LPC) SCALE AND MAGNITUDE ESTIMATION AND CASE III SCALING PROCEDURES
Ca. Schriesheim et al., THE EQUAL-INTERVAL NATURE OF SEMANTIC DIFFERENTIAL SCALES - AN EMPIRICAL-INVESTIGATION USING FIELDERS LEAST PREFERRED COWORKER (LPC) SCALE AND MAGNITUDE ESTIMATION AND CASE III SCALING PROCEDURES, Educational and psychological measurement, 54(2), 1994, pp. 253-262
Citations number
17
Categorie Soggetti
Psychology, Educational","Psychologym Experimental","Mathematical, Methods, Social Sciences
An assumption made in using semantic differential measures is that the
adjective pairs used are truly bipolar; otherwise, statistical analys
es may yield distorted findings. This article examines the 18 bipolar
adjective pairs used in Fiedler's Least Preferred Coworker (LPC) instr
ument by using two scaling techniques and a sample of 113 respondents.
To replicate and extend previous research, the 36 LPC adjectives were
ranked according to their relative interpersonal favorability. The sa
me sample also scaled the 36 adjectives using Stevens's magnitude esti
mation technique (MET). Both pair-comparison treatment of complete ran
ks scaling (under Thurstone's Case III) and the magnitude estimation p
rocedure indicated that true bipolarity did not characterize most of t
he paired adjectives. Implications for future LPC and semantic differe
ntial research are discussed.