Rotter Incomplete Sentences Blank, Second Edition

Purpose: Designed "as a screening instrument of overall adjustment."

Population: College students, adults, high-school students.

Score: Index of Overall Adjustment.

Time: (20-40) minutes.

Authors: Julian Rotter, Michael Lah, and Janet Rafferty.

Publisher: The Psychological Corporation.

Description: The Second Edition of the Rotter Incomplete Sentences Blank (RISB) is a projective measure of maladjustment with a semi-objective scoring system. This revised instrument provides direct information on personality conflicts. As scoring depends on intuitive clinical insights, cognizance of personality dynamics is essential for accurate interpretation. Although responses can also be scored qualitatively for projected motivational needs, as a general rule, interpretation of subjective scales is notoriously unreliable.

Scoring: Responses are rated on a 7-point ordinal scale (higher scores suggesting greater maladjustment) on the basis of omissions and incomplete responses, conflict responses, positive responses, or neutral responses. Overall scores generally range from 80 to 205 (on a scale from zero through 240). However, because of the diversity among individuals’ idiosyncratic responses, the RISB cannot readily be computer scored. This inevitably raises questions as to the objectivity and scoring consistency of RISB responses.

Reliability: Several studies reported in the RISB manual suggest uncertain reliability. Stability coefficients are reported as ranging from a low .38 (retest interval of 3 years) up to .82 (retest after only 1-2 weeks), so it cannot be assumed that the instrument is always reliable. Some inconsistency in scoring is evident because although interscorer reliabilities range from .72 to .99, item-rater reliabilities range from .44 to .93. Because the scoring examples are based solely on college samples, there is the further question as to their applicability for high school and adult groups. A positive feature of this test is its item homogeneity. The manual reports split-half estimates ranging form .74 to .86, and a Cronbach alpha coefficient of .69. This moderate level of item homogeneity suggests little item redundancy, and yet sufficient internal consistency to justify its use.

Validity: The RISB relies predominantly on face validity, so that item responses are readily amenable to distortion, depending on respondents’ lack of self-insight, and their conscious and unconscious motives. The instrument may act more effectively as a trait measure than as a state indicator of changes over time, and not be situationally sensitive. Despite these limitations, the RISB has received widespread use in both clinical and nonclinical settings.

Norms: College students.

Suggested Uses: Recommended uses of the RISB include screening, tracking changes in scores over time, group comparisons, and research.