Psychometrics of A New Version of the World View Survey

Thanks to the work of Gil Skidmore, M.Ed., data has been collected which allowed a major revision and refinement to be made in the World View Survey, a measure designed to assess important beliefs that may be changed by traumatic experience (Skidmore a nd Fletcher, 1997). Nearly 300 college students completed a preliminary 75-item version of the Survey, along with a measure of lifetime exposure to stressful events, the Young Adults Upsetting Times Checklist; a measure of PTSD, the Impact of Events Scal e (IES; Horowitz, Wilner, & Alvarez, 1979); and a measure of general psychopathology, the Brief Symptom Inventory (BSI; Derogatis & Melisaratos, 1983).

After submitting the 75 original items to item analysis, those that did not contribute to the overall scale were dropped. Others that were nearly duplicates of other items, with only slightly different wording, but that did not perform as well as thei r duplicates were also dropped. The remaining items were submitted to principal components factor analysis. Based on the scree plot of eigenvalues, nine factors were extracted. Items that did not load at least .35 on any of the nine factors were then d ropped. The remaining 50 items were again submitted to factor analysis, and nine factors were again extracted, using Varimax to derive orthogonal (fairly uncorrelated) factors.

In other words, the final 50 items were divided into nine subscales, named as follows: 1) Anxious Uncertainty (exemplified by beliefs such as, "Life does not seem to make much sense anymore"); 2) Inadequacy (e.g., "I am a jinx"); 3) Dangerous World (e. g., "The world is a dangerous place to live"); 4) Self-abnegation (e.g., "Sometimes I think I am not a very good person"); 5) Lack of Control (e.g., "I feel like I have control over my life" - if disagree); 6) Poor Ego-strength (e.g., "Since I have lived through some bad times, I have a better idea of what is important to me and what is not"- if disagree); 7) Poor Attachment (e.g., "It is easy for me to make friends" - if disagree); 8) Lack of Personal Empowerment (e.g., "I feel like nothing can keep me f rom getting what I want out of life anymore" - if disagree); and 9) Negative Outlook (e.g., "Nowadays I feel like every new day I am alive is a gift"- if disagree).

A higher-order factor analysis was conducted next, entering scores on the nine subscales into the analysis. This resulted in two second-order factors. The first factor included the following subscales: Anxious Uncertainty (factor loading = .91), Inad equacy (.86), Dangerous World (.69), Self-abnegation (.77), Lack of Control (.67), and Poor Attachment (.61). The second factor included the following subscales: Poor Ego-strength (loading = .83), Poor Attachment (.51), Lack of Empowerment (.53), and Neg ative Outlook (.79).

Only Poor Attachment loaded on both second-order factors. Ordinarily, it would be assigned to either both factors or to the one it loaded the highest on. However, because the purpose of the scales is to assess beliefs that may be affected by traumati c experiences, intercorrelations of the subscales with traumatic reactions, as indicated by scores on the IES were taken into consideration. All of the subscales included in the first second-order factor, except Poor Attachment, correlated significantly with the IES scale. Therefore, the first five subscales were combined to form a Trauma Reactive subscale, and the last four were combined to form a Negative Beliefs subscale. It appears that the subscales of the Trauma Reactive half of the World View Su rvey capture three of the four major beliefs that the item pool was originally meant to capture - Anxious Uncertainty beliefs and Lack of Control beliefs relate to beliefs about the uncertainty of the world, Inadequacy beliefs and Self-abnegation beliefs relate to beliefs about personal self-esteem, Dangerous World beliefs reflect a belief in the lack of safety in the world. In addition, the Poor Attachment subscale includes items designed to capture a belief in the worthiness of trust in others. The ot her three subscales in the Negative Beliefs second-order factor consist of items originally devised primarily to assess positive beliefs that some people have found to be associated with traumatic experiences, although they are now scored in a negative di rection.

For more information and a copy of the World View Survey, contact Ken Fletcher either by E-mail or write to him at Kenneth Fletcher, Ph.D., Psychiatry, University of Massachusetts Medical School, 5 5 Lake Avenue North, Worcester, MA 01655, USA.

Skidmore, G.L., & Fletcher, K.E. (November, 1997). Assessing Trauma's Impact on Beliefs: The World View Survey. A paper presented at the Thirteenth Annual Meeting of the International Society for Traumatic Stress Studies, Montreal, Quebec, Canada.

Table of Contents

Disclaimer: This is an unofficial UMMS page.