Revised Children's Manifest Anxiety Scale

Score: Physiological Anxiety, worry-Oversensitivity, Social concerns-concentration, Total Anxiety, Lie

Description: The Revised Children’s Manifest Anxiety Scale (RCMAS) is a self-report instrument designed to measure anxiety for children and adolescents aged 6-9 years. For children over 9 and a half years of age, it can be administered in a group situation. For first and second graders the examiner should read the items to the child. There are 37 items each of which requires a yes or no answer. The RCMAS was developed in 1978 to address criticisms of the original Children's Manifest Anxiety Scale (CMAS). Goals for revision of the scale were to (a) create an objective measure of children's anxiety suitable for group administration; (b) keep administration time to the minimum required for accurate, valid assessment; (c) make the reading level of items suitable for elementary school students but yet allow for use throughout the school years; (d) cover new areas of anxiety and determine whether anxiety would best be treated as unidimensional or multidimensional; (e) increase norms and information for diverse groups of children; and (f) assure that all items are good test items.

Scoring: The Total Anxiety score is based upon 28 items with 9 items comprising the Lie Scale. The Total Anxiety score and the Anxiety subscale scores are determined by the number of "yes" responses to the anxiety items. The Lie Score is determined by "yes" responses to the Lie subscale items and is used to determine if the child was making a valid attempt to respond. The three anxiety subscales should be interpreted cautiously and should be used only as an aid in hypothesis generation due to limited reliability levels. The Total Anxiety Score is expressed as a T score (M=50, sd=10) and the subscales are expressed as scale scores (M=10, sd = 3). Percentile ranks are provided for each of the RCMAS scores. Norms are provided at 1-year intervals and for each ethnic-sex combination for blacks and whites. The Lie Scale is a positive feature of the instrument and is designed to detect acquiescence, social desirability, or faking of responses.

Reliability: Reliability estimates for internal consistency for white and black males and females for each of the 12 age levels ranged from .42 (Age 6, black females) to .87 (Age 12, black males and Age 15, black males). Reliability estimates collapsed across the 12 age levels ranged from .79 to .85 (median=.82). Of the 48 coefficient alphas reported across age, race, and sex, 17 alphas fall below .80, but in general the Total Anxiety score is reliable enough across most ages to permit reliable interpretations. It should be noted however, that for black females; at ages 6, 8, 10, and 11, reliability coefficients are significantly lower than for white females at the same ages, and the authors urge that special care be taken when using the RCMAS with black females under 12 years of age.

Coefficient alphas are also reported for each of the three anxiety subscales and the Lie Scale. Most of these alphas fall below .80, some far below (.15) indicating poor domain sampling. Individual anxiety subscales are too unreliable to merit interpretation for most persons in the standardization sample. One study reported in the manual found a stability coefficient of .68 for the Total Anxiety score and .58 for the Lie Scale score for 534 children tested 9 months apart. Stability estimates increase dramatically when the interval between testing is shorter. All stability estimates reported in the manual are for children below grade 7. No stability data are reported for junior and senior high school students.

Validity: The manual reports higher correlations between the RCMAS Total Anxiety score and a measure of trait anxiety (r=.67) than between Total Anxiety and a measure of state anxiety (r=.10) despite the questions about its measure of trait anxiety raised by the low test-retest stability coefficients. Other than this, the validity data is sparse and the factors appear theoretically unrelated. Empirically, the correlation between the RCMAS factors range from .49 to .85 (median=.67), suggesting correlated factors and an oblique solution. However, the factor analysis of the instrument appears to be flawed. The standardization data should be refactored before test users can have faith in the dimensions underlying the RCMAS.

Norms: The RCMAS was standardized on 4,972 children between the ages of 6 and 19 years. The standardization sample contained 2,208 white males (44%), 2,176 white females (44%), 289 black males (5.8%), and 299 black females (6%). The sample was obtained from children in 13 states and more than 80 school districts from all major geographic regions of the U.S. The sample appears to be representative of the U.S. population, and the authors believe that "an excellent cross-section of school children was obtained" even though the manual provides no data comparing it to census data. Approximately 600 children in the sample were classified as handicapped (i.e., educably mentally retarded, learning disabled, or gifted). Users of the instrument can, however, be reasonably confident the RCMAS has a well-standardized and representative sample.

Suggested use: With future work with the scale, particularly the factor analysis undertaken the RCMAS may develop into one of the better measures of anxiety in children and adolescents.