Effect of number of response options in psychometric properties of four psychosocial scales
Abstract
Psychosocial scales are used profusely in studies in many different fields. However, psychometric research has not been conclusive regarding the number of response options that is optimal in measuring a construct. The objective of the present study was to compare the psychometric properties of four psychosocial scales using different number of response options. The participants (757 university students) were assigned to one of three independent groups to answer a battery consisting of four short scales that evaluated gratitude, optimism, subjective well-being and loneliness, with five, six or seven response options. A positive association was found between number of options and psychometric quality of the scales. The greatest variability (dispersion) was observed in versions of six and seven options; also, in these versions the internal consistency indices and the explained variance per-centages were the highest. It was concluded that the use of six and seven response options, specifically at short and one-dimensional psychometric scales, offers more precise measurements, compared to five. The implications of these findings are discussed.
Keywords
