Dataset Information

The value of suppressor effects in explicating the construct validity of symptom measures.

ABSTRACT: Suppressor effects are operating when the addition of a predictor increases the predictive power of another variable. We argue that suppressor effects can play a valuable role in explicating the construct validity of symptom measures by bringing into clearer focus opposing elements that are inherent--but largely hidden--in the measure's overall score. We illustrate this point using theoretically grounded, replicated suppressor effects that have emerged in analyses of the original Inventory of Depression and Anxiety Symptoms (IDAS; Watson et al., 2007) and its expanded 2nd version (IDAS-II; Watson et al., 2012). In Study 1, we demonstrate that the IDAS-II Appetite Gain and Appetite Loss scales contain both (a) a shared distress component that creates a positive correlation between them and (b) a specific symptom component that produces a natural negative association between them (i.e., people who recently have experienced decreased interest in food/loss of appetite are less likely to report a concomitant increase in appetite/weight). In Study 2, we establish that mania scales also contain 2 distinct elements-namely, high energy/positive emotionality and general distress/dysfunction-that oppose each another in many instances. In both studies, we obtained evidence of suppression effects that were highly robust across different types of respondents (e.g., clinical outpatients, community adults, college students) and using both self-report and interview-based measures. These replicable suppressor effects establish that many homogeneous, unidimensional symptom scales actually contain distinguishable components with distinct--at times, even antagonistic--properties.

SUBMITTER: Watson D

PROVIDER: S-EPMC3780394 | biostudies-literature | 2013 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

The value of suppressor effects in explicating the construct validity of symptom measures.

Watson David D Clark Lee Anna LA Chmielewski Michael M Kotov Roman R

Psychological assessment 20130624 3

Suppressor effects are operating when the addition of a predictor increases the predictive power of another variable. We argue that suppressor effects can play a valuable role in explicating the construct validity of symptom measures by bringing into clearer focus opposing elements that are inherent--but largely hidden--in the measure's overall score. We illustrate this point using theoretically grounded, replicated suppressor effects that have emerged in analyses of the original Inventory of De ...[more]

PMID: 23795886

Similar Datasets

Project description:BACKGROUND:Various self-reported or clinician-reported (as a proxy) measures exist to quantify the burden of schizophrenia on patients. Evidence of the psychometric relationship between these measures to inform their practical use is limited. OBJECTIVES:Our objective was to conduct an exploratory analysis of the construct validity of patient-reported (EQ-5D, SF-6D, WEMWBS, SQLS subscales of Psychosocial, Motivation, Symptoms) versus clinician-reported measures (PANSS, CGI-SCH, NSA-4, HoNOS-PbR) to inform future use of patient-reported measures for burden-of-illness assessment and/or economic evaluation. METHODS:In an adult patient population with schizophrenia, construct validity was assessed in relation to convergent and known-group validity. Convergent validity was assessed using Spearman's rank absolute correlation strength (ACS: weak ≤ 0.3, moderate = 0.3 < 0.5, strong ≥ 0.5) and graphically using locally weighted scatterplot smoothing (LOWESS) techniques. Known-group validity was assessed using Cohen's d absolute effect size (AES: small ≤ 0.5, moderate = 0.5 < 0.8, large ≥ 0.8). Floor and ceiling effects were assessed as a proxy of sensitivity in this cross-sectional study. Statistical significance was assessed at the 5% threshold level (p < 0.05). Across head-to-head assessments, the frequency of producing the strongest ACS, largest AES, and statistically significant results determined the best overall construct validity. RESULTS:Overall, 304 patients consented to the study. In relation to statistically significant results, the SF-6D most frequently exhibited the strongest ACS and largest AES against the clinician-reported measure scores (ACS range 0.084-0.436; AES range 0.043-0.746), and the SQLS Motivation subscale most frequently exhibited the weakest/smallest values (ACS range 0.009-0.157; AES range 0.002-0.397), although these results were mixed according to the clinician-reported measure used for comparative analysis (ACS range 0.009-0.529; AES range 0.002-0.934). CONCLUSION:The SF-6D indicated the best (mostly moderate) construct validity but still missed the negative symptoms of the condition. Although further evidence is required to confirm or refute these exploratory results, compared with the EQ-5D, the SF-6D can be self-reported to better capture generic health-related quality-of-life aspects of schizophrenia for the purpose of economic evaluation. The lack of construct validity for SQLS Motivation and Symptoms subscales were hypothesized post-hoc to be representative of the complementary information elicited by the subscales not captured by the clinician-reported measures. Therefore, the SQLS can be self-reported to capture complementary (i.e., additional) information relative to clinician-reported measures.

Project description:BackgroundInflammatory Conditions of the Lactating Breast (ICLB) affect more than one in five lactating mothers, yet no fully validated outcome measures exist to aid clinicians in their patient-centred care of women with ICLB. The Breast Inflammatory Symptom Severity Index (BISSI) is an ICLB-specific clinician administered patient-reported outcome measure, currently used by Australian clinicians, who treat mothers with ICLB. To date the BISSI has undergone partial psychometric development. This study, therefore, aimed to undertake the next stage of psychometric development by determining the construct validity and internal consistency of the BISSI.MethodsA retrospective audit was conducted on patient records of 160 mothers who were treated for ICLB, at a private physiotherapy practice in Melbourne, Australia. An electronic data capture tool was used to collate BISSI scores and associated ICLB assessment variables. Construct validity was determined through factor analysis and discriminant performance. Reliability was determined by assessing measures of internal consistency.ResultsFactor analysis established that BISSI items (n = 10) loaded on to four factors, Wellness, Pain, Physical Characteristics of Affected Area (PCAA), and Inflammation, which together, explained 71.2% of variance. The remaining item ('Wellness/sickness unspecified') did not load. Wellness, Pain, PCAA and Inflammation factors individually and collectively displayed the ability to discriminate symptom severity, as scores were significantly higher in mothers with high symptom severity (assessed via AUC close to or >0.7 and P value <0.005 for each factor). The BISSI demonstrated internal consistency with an overall Cronbach alpha of 0.742.ConclusionsThe BISSI has adequate construct validity, demonstrating behaviour consistent with theoretical constructs of inflammation severity, via its dimensionality and ability to discriminate symptom severity. The BISSI also has adequate internal consistency demonstrating reliability. Therefore, clinicians can have confidence that the BISSI is valid, the individual item scores are correlated, and the concepts are consistently measured.

Project description:BackgroundThe evaluation of treatment efficacy for individuals with fragile X syndrome (FXS) or intellectual disability (ID) more generally has been hampered by the lack of adequate outcome measures. We evaluated expressive language sampling (ELS) as a procedure for generating outcome measures for treatment research in FXS. We addressed: (a) feasibility, (b) practice effects over two administrations, (c) test-retest reliability over the repeated administrations, and (d) construct validity. We addressed these issues for the full sample as well as for subgroups defined by age, IQ, and ASD status.MethodsParticipants were 106 individuals with FXS between ages 6 and 23 years who had IQs within the range of intellectual disability (IQ < 70). ELS procedures for collecting samples in conversation and narration were followed and analyzed separately. Five measures were derived from transcripts segmented into C-units (i.e., an independent clause and its modifiers): number of C-units per minute (talkativeness), number of different word roots (vocabulary), C-unit length in morphemes (syntax), percentage of C-units containing dysfluency (utterance planning), and percentage of C-units that were fully or partly unintelligible (articulatory quality). ELS procedures were administered twice at 4-week intervals for each participant. Standardized tests and informant reports were administered and provided measures for evaluating construct validity of ELS measures.ResultsWe found low rates of noncompliance, suggesting the task can be completed meaningfully by most individuals with FXS, although noncompliance was higher for younger, lower IQ, and more autistic participants. Minimal practice effects and strong test-retest reliability over the 4-week interval were observed for the full sample and across the range of ages, IQs, and autism symptom severity. Evidence of convergent construct validity was observed for the measures of vocabulary, syntax, and unintelligibility for the full sample and across the range of IQ and autism symptom severity, but not for participants under age 12. Conversation and narration yielded largely similar results in all analyses.ConclusionsThe findings suggest that the ELS procedures are feasible and yield measures with adequate psychometric properties for a majority of 6 to 23 years with FXS who have ID. The procedures work equally well regardless of level of ID or degree of ASD severity. The procedures, however, are more challenging and have somewhat less adequate psychometric properties for individuals with FXS under age 12.

Project description:[Correction Notice: An Erratum for this article was reported online in Psychological Assessment on Nov 19 2018 (see record 2018-60122-001). In the article, the affiliations of multiple authors were incorrectly listed in the author byline and author note. For Daniel J. Lee, Michelle J. Bovin, Denise M. Sloan, Terence M. Keane, and Brian P. Marx, the affiliations should have read "National Center for PTSD, Boston, Massachusetts; VA Boston Healthcare System, Boston, Massachusetts; and Department of Psychiatry, Boston University School of Medicine." For author Paula P. Schnurr, the affiliations should have read "National Center for PTSD, White River Junction, Vermont; Department of Psychiatry, Geisel School of Medicine at Dartmouth." All versions of this article have been corrected.] The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) 4-factor model of posttraumatic stress disorder (PTSD) has demonstrated adequate fit in several confirmatory factor analysis (CFA) studies. Although several alternative measurement models have demonstrated better fit, there is no consensus yet on the best model, and newly proposed models lack sufficient construct validation. Notably, these studies have relied exclusively on questionnaire data, and thus their findings may be attributable to a method effect. This study examined the factor structure of DSM-5 PTSD symptoms using both questionnaire and interview data to determine the impact of assessment method on factor structure and construct validity of alternative model symptom clusters. Participants (N = 380) were veterans who completed the PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013) and the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5; Weathers et al., 2013). Fit was similar across models. However, the seven-factor Hybrid model (Armour et al., 2015) fit best. Limited evidence of a method effect was observed. Results of construct validity analyses were mixed; some of the newly proposed symptom clusters demonstrated hypothesized differential associations with external correlates, but others did not. These findings suggest that results of previous DSM-5 PTSD CFAs supporting the Hybrid model are not attributable to a method effect. However, observed limited difference in model fit and mixed construct validity evidence raise concerns regarding the value of parsing DSM-5 symptom clusters. Constructs implied by the new factors in the more complex measurement models of PTSD require greater explication and construct validation. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Dataset Information

The value of suppressor effects in explicating the construct validity of symptom measures.

Publications

The value of suppressor effects in explicating the construct validity of symptom measures.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets