Note on two generalizations of coefficient alpha

An important relationship is given for two generalizations of coefficient alpha, Rajaratnam, Cronbach and Gleser's generalizability formula for stratified-parallel tests and Raju's coefficient beta.The author gratefully acknowledges the generous assistance given by reviewers and the editor.     

Ordinal data: An alternative distribution

To date, virtually all techniques appropriate for ordinal data are based on the uniform probability distribution over the permutations. In this paper we introduce and examine an alternative probability model for the distribution of ordinal data. Preliminary to deriving the expectations of Spearman's rho and Kendall's tau under this model, we show how to compute certain conditional expectations of rho and tau under the uniform distribution. The alternative probability model is then applied to ordinal test theory, and the calculation of true scores and test reliability are discussed.     

Test-retest reliability of retrospective self-reports in three populations of alcohol abusers

The reliability of alcohol abusers' retrospective self-reports was evaluated using a time-line follow-back interview technique. Independent groups of male subjects were interviewed in three different treatment settings (outpatient,n=12; inpatient,n=12; residential,n=12) on two separate occasions to assess the test-retest reliability of their self-reports of daily drinking and related events occurring 1 year prior to treatment. Correlational and scatterplot analyses generally showed a high correlation and low discrepancy between the two sets of interview responses for all groups. However, since some population differences were found in the reliability of reports for some variables, further research is needed to delineate the conditions under which reliable and valid retrospective self-reports can be obtained from different populations of alcohol abusers.This research was supported, in part, by Grant No. AA07072 from the National Institute on Alcohol Abuse and Alcoholism to Vanderbilt University.     

Traits,states, situations,and uncertainty

The shift of paradigm from psychodynamic therapy to behavior modification has changed the views of assessment and challenged traditional broad trait concepts. Behavioral assessment has used narrow, situation-specific trait tests, state self-report tests given in situations, and behavioral observations and performance ratings. Comparison of these types of measures are reported from a study of fear reactions in three situations. Narrow trait measures are generally more predictive of behavior than broad trait measures. State measures are even more predictive when given just before the performance. But only a sampling of such state measures can be used to define a trait because of the lower reliability of states. It is argued that behavior in situations is only predictable when an adequate number of behavioral samples is used.     

Clinical Frequencies Recording System: Standardizing staff observations by event recording

Certain aspects of staff performance and resident behavior are less adequately assessed by standardized time sampling procedures than by complete recording of the incidence of specified events. An approach to gathering such data through systematizing the observations of clinical staff — the Clinical Frequencies Recording System — is described. Scores from the Clinical Frequencies System reflect the rate per opportunity with which individual behaviors are performed, with computer summarization providing higher-level behavioral indexes for research and evaluation purposes. Staff training and system implementation are discussed, and reliability and validity data from a system employed on two different treatment units over a period of several years are presented.The research reported in this article was supported, in part, by Public Health Service Grants MH-15553 and MH-25464 from the National Institute of Mental Health.Presented at the 87th Annual Meetings of the American Psychological Association, New York City, September 1979, as part of a symposium on New assessment systems for residential treatment, management, research, and evaluation.     

MMCPC在中学生中的应用

为了解 MMCPC在中国中学生中是否适用 ,本研究选取了浙江省1 870名中学生进行测试 ,检验 MMCPC的信度和效度。结果显示 ,MMCPC在中国中学生群体中有较好的信度和一定效度。     

Features of Interview Questions Associated with Attenuation of Symptom Reports

Previous studies have suggested that discrepant reporting in a test–retest reliability paradigm is not purely random measurement error, but partly a function of a systematic tendency to say no during retest to questions answered positively at initial testing (attenuation). To examine features of interview questions that may be associated with attenuation, three raters independently assessed the structural and content features of questions from the Diagnostic Interview Schedule for Children (version 2.3) and linked these to data from a test–retest reliability study of 223 community respondents (parent and child reports). Results indicated that for both parent and youth reports, item features most strongly associated with attenuation were (a) being a stem question (asked of all respondents, regardless of any skip structure); (b) question placement in the first half of the interview; (c) question length; (d) question complexity; or (e) requiring assessment of the timing, duration, or frequency of a symptom. Findings may be explained by participants' conscious efforts to avoid further questions or by their learning more about the nature and purpose of the interview as they gain more experience; alternatively, findings may represent a methodological artifact of structured interview design.     

Interviewing as Communication: An Alternative Way of Administering the Diagnostic Interview Schedule for Children

Parents of 24 children referred to an outpatient psychology clinic (mean child age 10.8, range 6–15) were administered the Diagnostic Interview Schedule for Children Version 2.3 (DSIC-2.3) twice in a 1-week test–retest reliability design (mean retesting interval = 7.5 days, range = 6–11 days). An alternative mode of administration of the DISC, based on communication principles, was used, involving (a) a schematic representation of the areas to be covered; (b) definition of a common language for the categories, diagnoses, and criteria; and (c) the respondent being allowed to select the order in which the diagnostic areas were covered. The DISC items and modules were unchanged. Symptom scores derived from the DISC were highly reliable over 1 week (average ICC = .85, range = .67–.95) and showed no attenuation from Time 1 to Time 2. Reliability of DSM diagnoses averaged kappa = .80 (range = .63–1.0). There was no significant attenuation in diagnoses from Time 1 to Time 2. Overall, this alternative way of administering the DISC appears to have promise for reducing attenuation and boosting the reliability—and ultimately the validity—of child psychiatric diagnoses. Further investigations of the mechanisms underlying these effects, and further studies with child and adolescent respondents and nonreferred community samples are recommended.