How Does the Number of Response Options Impact the Psychometric Properties of the Rosenberg Self-Esteem Scale?

Psychometric benefits of self-chosen rating scales over given rating scales

Rating scales are susceptible to response styles that undermine the scale quality. Optimizing a rating scale can tailor it to individuals' cognitive abilities, thereby preventing the occurrence of response styles related to a suboptimal response format. However, the discrimination ability of individuals in a sample may vary, suggesting that different rating scales may be appropriate for different individuals. This study aims to examine (1) whether response styles can be avoided when individuals are allowed to choose a rating scale and (2) whether the psychometric properties of self-chosen rating scales improve compared to given rating scales. To address these objectives, data from the flourishing scale were used as an illustrative example. MTurk workers from Amazon's Mechanical Turk platform (N = 7042) completed an eight-item flourishing scale twice: (1) using a randomly assigned four-, six-, or 11-point rating scale, and (2) using a self-chosen rating scale. Applying the restrictive mixed generalized partial credit model (rmGPCM) allowed examination of category use across the conditions. Correlations with external variables were calculated to assess the effects of the rating scales on criterion validity. The results revealed consistent use of self-chosen rating scales, with approximately equal proportions of the three response styles. Ordinary response behavior was observed in 55-58% of individuals, which was an increase of 12-15% compared to assigned rating scales. The self-chosen rating scales also exhibited superior psychometric properties. The implications of these findings are discussed.

Brief Report: Does the Number of Response Options Matter for the BFI-2? Conceptual Replication and Extension

We evaluated how the number of response options affects the psychometric properties of the Big Five Inventory-2 (BFI-2). Using two large samples collected from a market research company (Ns = 893 and 1,213), we tested how different response options of the BFI-2 influenced scale score distributions, internal consistency estimates, convergent validity correlations, and criterion validity correlations. Results suggest that score distributions were impacted by the number of response options such that ceiling and floor effects were more common when using two or three response options than when using more options. Estimates of Cronbach's alpha were generally lower with fewer scale points as compared with more scale points, but these effects disappeared when ordinal alpha was used. There were no systematic effects of response options on convergent validity and criterion validity correlations. Given these results, there seems to be few psychometric reasons for deciding whether to administer personality items with five, six, or seven scale points.

Are the Steps on Likert Scales Equidistant? Responses on Visual Analog Scales Allow Estimating Their Distances

A recurring question regarding Likert items is whether the discrete steps that this response format allows represent constant increments along the underlying continuum. This question appears unsolvable because Likert responses carry no direct information to this effect. Yet, any item administered in Likert format can identically be administered with a continuous response format such as a visual analog scale (VAS) in which respondents mark a position along a continuous line. Then, the operating characteristics of the item would manifest under both VAS and Likert formats, although perhaps differently as captured by the continuous response model (CRM) and the graded response model (GRM) in item response theory. This article shows that CRM and GRM item parameters hold a formal relation that is mediated by the form in which the continuous dimension is partitioned into intervals to render the discrete Likert responses. Then, CRM and GRM characterizations of the items in a test administered with VAS and Likert formats allow estimating the boundaries of the partition that renders Likert responses for each item and, thus, the distance between consecutive steps. The validity of this approach is first documented via simulation studies. Subsequently, the same approach is used on public data from three personality scales with 12, eight, and six items, respectively. The results indicate the expected correspondence between VAS and Likert responses and reveal unequal distances between successive pairs of Likert steps that also vary greatly across items. Implications for the scoring of Likert items are discussed.

Development of an Inconsistent Responding Scale for the HEXACO Personality Inventory-Revised

Inconsistent or careless responding is a significant threat to the validity of self-reported personality data. Using archival samples of undergraduate and community participants, we developed an inconsistent responding scale using items that appear on both the 60- and 100-item versions of the HEXACO Personality Inventory-Revised-two widely used measures of the HEXACO model of personality trait structure. We identified pairs of correlated HEXACO items in Sample 1 and created a total inconsistent responding score by summing absolute differences between each item pair. The Brief Response Inconsistency Evaluation (BRIE) for the HEXACO effectively differentiated between genuine and randomly generated responses across samples. The BRIE also correlated as expected with other measures of careless responding and relevant personality traits (e.g., conscientiousness). Tentative cut scores for the BRIE that appear to provide a reasonable balance between sensitivity and specificity in Sample 1 were investigated. Future research should examine the BRIE with different populations and translations of the HEXACO inventories and further investigate the effectiveness of the recommended cut scores.