Is Truth Stranger Than Fiction: Symptom Clustering With Factor & Cluster Analysis

2.50
Hdl Handle:
http://hdl.handle.net/10755/157806
Type:
Presentation
Title:
Is Truth Stranger Than Fiction: Symptom Clustering With Factor & Cluster Analysis
Abstract:
Is Truth Stranger Than Fiction: Symptom Clustering With Factor & Cluster Analysis
Conference Sponsor:Western Institute of Nursing
Conference Year:2009
Author:Meek, Paula, RN, PhD, FAAN
P.I. Institution Name:University of Colorado Denver, Nursing
Title:Professor
Contact Address:13120 East 19th Ave, Aurora, CO, 80045, USA
Contact Telephone:303-724-2878
Co-Authors:Ellyn Matthews, Professor
Specific Aims: The overall aim of this presentation is to compare two methods of cluster formation, exploratory factor analysis, and cluster analysis, with the same data set and point out variations in assumptions, techniques, and results. A secondary aim is to suggest guidelines that can inform an analysis strategy aimed at cluster formation. Rationale: The literature dealing with symptom occurrence and management in recent years has been occupied with discussions of symptom clusters and their meaning and usefulness. A symptom cluster is defined as three or more coexisting symptoms that are related but may not share the same etiology. However, there is a long standing problem in understanding the occurrence and meaning of symptoms related to the use of diverse methodological approaches to analyzing the data. Few articles have sought to look at differences in symptom cluster formation by analysis strategy and methods used. Methods: All participants were male, with an average age of 39 (SD 8). The focus of these analyses will be the symptom scale that contained 23 different symptoms (e.g., loss of appetite, muscle pain) each rated as present or not present (0 or 1). Approximately 20% of individuals (n=180) denied having any of the symptoms and were eliminated from further analysis. Factor analysis was carried out using principal components extraction and varimax rotation. Hierarchical cluster analyses with different methods (within subjects, Ward, and centroid) were conducted and compared to the factor analysis results. Results: The mean number of symptoms reported was 6 (SD 4.6) with 47% of the sample reporting only 1 or 2 symptoms. Depression was reported the most (44%) with wheezing the least (8%). The factor analysis produced six factors that explained 51% of the variance (see table). These six factors were not consistent with the cluster analysis results. Factor/Cluster one was consistent across methods, with the best fit seen using the Ward method. Implications: The results show that different analysis techniques can result in symptom cluster dissimilarity. A conservative approach to determine symptom clusters must include use of multiple techniques to evaluate consistency and stability in the results. In addition, it is clear that there is a need for increased clarity of the decision rules and reasons for using a particular technique in developing symptom clusters that may inform investigations and care.
Repository Posting Date:
26-Oct-2011
Date of Publication:
17-Oct-2011
Sponsors:
Western Institute of Nursing

Full metadata record

DC FieldValue Language
dc.typePresentationen_GB
dc.titleIs Truth Stranger Than Fiction: Symptom Clustering With Factor & Cluster Analysisen_GB
dc.identifier.urihttp://hdl.handle.net/10755/157806-
dc.description.abstract<table><tr><td colspan="2" class="item-title">Is Truth Stranger Than Fiction: Symptom Clustering With Factor &amp; Cluster Analysis</td></tr><tr class="item-sponsor"><td class="label">Conference Sponsor:</td><td class="value">Western Institute of Nursing</td></tr><tr class="item-year"><td class="label">Conference Year:</td><td class="value">2009</td></tr><tr class="item-author"><td class="label">Author:</td><td class="value">Meek, Paula, RN, PhD, FAAN</td></tr><tr class="item-institute"><td class="label">P.I. Institution Name:</td><td class="value">University of Colorado Denver, Nursing</td></tr><tr class="item-author-title"><td class="label">Title:</td><td class="value">Professor</td></tr><tr class="item-address"><td class="label">Contact Address:</td><td class="value">13120 East 19th Ave, Aurora, CO, 80045, USA</td></tr><tr class="item-phone"><td class="label">Contact Telephone:</td><td class="value">303-724-2878</td></tr><tr class="item-email"><td class="label">Email:</td><td class="value">paula.meek@ucdenver.edu</td></tr><tr class="item-co-authors"><td class="label">Co-Authors:</td><td class="value">Ellyn Matthews, Professor</td></tr><tr><td colspan="2" class="item-abstract">Specific Aims: The overall aim of this presentation is to compare two methods of cluster formation, exploratory factor analysis, and cluster analysis, with the same data set and point out variations in assumptions, techniques, and results. A secondary aim is to suggest guidelines that can inform an analysis strategy aimed at cluster formation. Rationale: The literature dealing with symptom occurrence and management in recent years has been occupied with discussions of symptom clusters and their meaning and usefulness. A symptom cluster is defined as three or more coexisting symptoms that are related but may not share the same etiology. However, there is a long standing problem in understanding the occurrence and meaning of symptoms related to the use of diverse methodological approaches to analyzing the data. Few articles have sought to look at differences in symptom cluster formation by analysis strategy and methods used. Methods: All participants were male, with an average age of 39 (SD 8). The focus of these analyses will be the symptom scale that contained 23 different symptoms (e.g., loss of appetite, muscle pain) each rated as present or not present (0 or 1). Approximately 20% of individuals (n=180) denied having any of the symptoms and were eliminated from further analysis. Factor analysis was carried out using principal components extraction and varimax rotation. Hierarchical cluster analyses with different methods (within subjects, Ward, and centroid) were conducted and compared to the factor analysis results. Results: The mean number of symptoms reported was 6 (SD 4.6) with 47% of the sample reporting only 1 or 2 symptoms. Depression was reported the most (44%) with wheezing the least (8%). The factor analysis produced six factors that explained 51% of the variance (see table). These six factors were not consistent with the cluster analysis results. Factor/Cluster one was consistent across methods, with the best fit seen using the Ward method. Implications: The results show that different analysis techniques can result in symptom cluster dissimilarity. A conservative approach to determine symptom clusters must include use of multiple techniques to evaluate consistency and stability in the results. In addition, it is clear that there is a need for increased clarity of the decision rules and reasons for using a particular technique in developing symptom clusters that may inform investigations and care.</td></tr></table>en_GB
dc.date.available2011-10-26T20:13:21Z-
dc.date.issued2011-10-17en_GB
dc.date.accessioned2011-10-26T20:13:21Z-
dc.description.sponsorshipWestern Institute of Nursingen_GB
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.