Synopsis of the Derogatis Affects Balance ScaleDABS/DABS-SF
Derogatis, Leonard R.
Derogatis Stress Profile (DSP®)
Clinical Psychometric Research Inc.,
1228 Wine Spring Lane, Towson, MD 21204
Year Developed: 1975, 1995
Primary Measurement Constructs: Principal dimensions of positive and negative emotional experience, with a particular emphasis on affects balance.
The DABS is a multidimensional self-report mood and affects inventory comprised of 40 adjective-items. The DABS measures affectivity and affects balance via 8 primary affects dimensions and five global scores. The positive affects dimensions are labeled joy, contentment, vigor and affection. The negative dimensions are anxiety, depression, guilt and hostility. The DABS global scores consist of the Positive Total score (PTOT), Negative Total score (NTOT), and the Affects Balance Index (ABI). Recently, two additional globals, the Affects Expressiveness Index (AEI) and the Positive Affects Ratio (PAR), were developed, and a brief form of the scale, the DABS-SF, has been normed and introduced.
The primary dimensions of the DABS are designed to reflect the principal components of positive and negative emotional experience, while the five global scores are intended to communicate an overall or summary picture of the respondent’s emotional status and well-being. The Positive Affects Total (PTOT) is defined as the sum of all scores on the four positive affects dimensions of joy, contentment, vigor and affection. Similarly, the Negative Affects Total (NTOT) is represented as the sum of scores on the four negative dimensions of anxiety, depression, guilt and hostility. The Affects Balance Index (ABI) is algebraically defined as (PTOT-NTOT)/ 20, and conveys the degree of balance or symmetry between positive and negative emotional experiences. The Affects Expressiveness Index (AEI) is defined as the sum total of affective expression, regardless of valence (i.e., regardless of positive or negative direction). It illustrates the individual’s affective “charge” or total level of affectivity. The Positive Affects Ratio (PAR) illustrates a distinct approach to measuring global affective status. It is defined as the ratio of positive affectivity to total affectivity (i.e., positive plus negative affectivity) on the DABS. The PAR reflects the proportion of affective experience represented as positive emotion.
Norms and Standardized Scores
Norms have been developed for both the DABS and the DABS-SF. In the case of the former, the norm is based on approximately 500 community respondents. The DABS-SF norm is currently based on approximately 300 community individuals. On both versions of the test, slight gender differences have been found to exist on several DABS measures. They are minimal, however, and judged as clinically insignificant, thereby avoiding the necessity for gender-keyed norms.
Both DABS and DABS-SF norms are represented as standardized scores in terms of Area T-scores. The Area transformation possesses distinct advantages over Linear transformations in that the former provide accurate percentile equivalents (i.e., T-score of 30=2nd centile; T-score of 40=16th centile; T-score of 50=50th centile; T-score of 60 = 84th centile; T-score of 70=98th centile, etc.). This important characteristic of area-transformations is only true of linear-transformations when the underlying raw score distributions are perfectly normal.
Clinical/Research Instrument, Self-Report
Arabic, Chinese, English, French, French Canadian, Icelandic, Indian, Korean, Norwegian, Spanish and Turkish.
The DABS is comprised of 40 single-word, adjective items which describe various positive and negative mood states. Each adjective is rated by the respondent relative to his/her recent personal experience on 5-point Likert-type scales, ranging from 0=”never” to 4=”always”. The DABS requires approximately 3-5 minutes to complete under typical circumstances. The DABS-SF is comprised of 20 adjective items represented in the same
format and requires approximately 2-4 minutes to complete. The DABS-SF is scored only in terms of the 5 global scores.
Reliability and Validity
As has been pointed out numerous times in various texts on psychological measurement, reliability has to do with the precision or accuracy of measurement, while validity addresses the essence of the attribute under assessment. Demonstrations of the former are typically established through one or more highly specific sets of operations, while confirmation of the latter is much more programmatic, requiring multiple experiments and empirical studies. Viewed from one perspective, the validation process is unending, since the limits of generalizability (i.e., valid application) of a substantive psychological measuring instrument are constantly being tested, expanded and refined.
Reliability of the DABS: In the case of self-report inventories, there are usually two essential forms of reliability: internal consistency reliability, and test-retest. Internal consistency reliability is a measure of homogeneity or consistency of item selection; it reflects the degree to which the items chosen to operationalize the test or subtest in question are drawn from the same population of items. Test-retest reliability is a measure of temporal stability; it demonstrates the degree to which scores achieved at a particular time of assessment correlate with scores achieved on subsequent assessment occasions.
Both internal consistency and test retest coefficients have been empirically established for the DABS and are given in detail in Derogatis & Rutigliano (1996). In summary, coefficients alpha (á) based on a sample of 355 psychiatric inpatients, ranged from a low of .79 for anxiety to a high of .85 for depression, vigor and affection. Global coefficients were in the low to mid .90’s. Test-retest coefficients (rtt) were reported based on a small sample of 16 primary breast cancer patients, with one week separating the two assessment occasions. Coefficients for the primary DABS dimensions ranged from a low of .79 for hostility to a high of .84 for contentment; stability coefficients for the globals were all in the low .80’s.
Validation of Dimensional Structure: An important demonstration in the construct validation of a multidimensional psychological test involves the verification of dimensional structure. The operational definitions (i.e., dimensions) of the primary hypothesized constructs measured by the test require confirmation in terms of real world data. In the case of the DABS, validation requires verification at two distinct levels of conceptual structure: At a fundamental level, the 8 hypothesized primary affects dimensions, and at a broader conceptual level, the overarching constructs of positive and negative affectivity. To achieve verification of dimensional structure, admissions DABS from 355 psychiatric inpatients at a private psychiatric hospital were scored and subjected to principal components analysis with an orthogonal varimax rotation. Six factors or dimensions were recovered which met the Scree criterion and accounted for approximately 70% of the variance in the correlation matrix. Three factors emerged which clearly reflected positive affects dimensions: a combined joy/contentment dimension, an affection dimension, and a vigor dimension. On the negative side, the largest factor represented a combined depression/guilt factor, with two additional components clearly reflecting anxiety and hostility. Although not a perfect recovery of the hypothesized 8-factor structure, the analysis did confirm the majority of hypothesized constructs. In order to verify the higher-order factors of positive and negative affectivity, the primary dimension correlation matrix was also subjected to principal components analysis with orthogonal varimax rotation. Two principal components were identified which accounted for 72.2% of the variance. The factor structure derived from this analysis clearly identified positive and negative affectivity as the two superordinate dimensions measured by the DABS (Derogatis & Rutigliano, 1996).
Although non-dimensional, it is worth noting that the five global scores of the DABS-SF correlate from between .94 to .97 with corresponding measures from the DABS. Predictive Validity: When most clinicians and researchers speak about the validity of a psychological outcome measure they are usually referring to its predictive validity. Predictive validity addresses the practical utility of the test: how well, for example, it identifies positive cases, discriminates effective from ineffective treatments, distinguishes treatment responders from non-responders, or correlates with other independent measures of outcome. The DABS has proven highly effective as a primary outcome measure in all of these clinical research contexts. In the citations that follow, the DABS has demonstrated high effects sensitivity in such diverse contexts as psychosocial oncology, coping with loss, stress paradigms, pharmacotherapy outcomes, psychotherapy outcomes, affective disorders and sexual dysfunctions.
How to Obtain
The DABS and DABS-SF are distributed exclusively by
Clinical Psychometric Research Inc.
1228 Wine Spring Lane
Towson, MD 21204.
Phone 1-800-245-0277; 1-(410) 321-6165; FAX 1-(410) 321-6341.
Leonard R. Derogatis, Ph.D.
Derogatis, LR (1996) Derogatis Affects Balance Scale (DABS): Preliminary Scoring, Procedures & Administration Manual. Baltimore, MD, Clinical Psychometric Research.
Derogatis, LR & Rutigliano, PJ (1996) Derogatis Affects Balance Scale (DABS). In B. Spilker (Ed.) Quality of Life and Pharmacoeconomics In Clinical Trials, 2nd Ed. Lippincott-Raven, Philadelphia, PA.
Derogatis LR, Abeloff MD & Melisaratos, N. (1979) Psychological coping mechanisms and survival time in metastatic breast cancer. JAMA .242: 1504-1508.
Derogatis LR & Meyer JK. (1979) A psychological profile of the sexual dysfunctions. Archives of Sexual Behavior. 8, 201-223.
Garamoni GL, Reynolds CF, Thase ME, Frank, E & Fasiczka AL.(1992) Shifts in affective balance during therapy of major depression. J Consulting & Clinical Psychology 60, 260-266.
Rabins PV, Fitting MD, Eastham J & Fetting J. (1990) The emotional impact of caring for the chronically ill. Psychosomatics. 31, 331-336.
Watson D & Tellegen A. (1985) Toward a consensual structure of mood. Psychological Bulletin, 98, 219-235.