SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Gothwal VK, Sumalini R, Irfan SM, Giridhar A, Bharani S. Optom. Vis. Sci. 2013; 90(8): 828-835.

Affiliation

*PhD, FAAO †BSOpt Kallam Anji Reddy Campus, L. V. Prasad Eye Institute (VKG); Meera and L. B. Deshpande Centre for Sight Enhancement, Vision Rehabilitation Centres (RS, SB); and Bausch and Lomb School of Optometry, L. V. Prasad Eye Institute (SMI, AG), Hyderabad, Andhra Pradesh, India.

Copyright

(Copyright © 2013, Lippincott Williams and Wilkins)

DOI

10.1097/OPX.0b013e3182959b52

PMID

23792363

Abstract

PURPOSE: To explore the psychometric properties of the revised Olweus Bully/Victim Questionnaire (OBVQ) in children with visual impairment (VI) using Rasch analysis.

METHODS: One hundred fifty Indian children with VI between 8 and 16 years (mean age, 11.6 years; 69% male; mean acuity in the better eye of 0.80 logMAR [Snellen, 20/126]) were administered the revised OBVQ. The 40-item revised OBVQ was developed to assess victimization (i.e., being bullied) and bullying (bullying others) in normally sighted schoolchildren. Only 16 items are used for Rasch analysis and are divided into two parts: I (victimization, eight items) and II (bullying others, eight items). Separate Rasch analysis was conducted for both parts, and the psychometric properties investigated included behavior of rating scale, extent to which the items measured a single construct (unidimensionality by fit statistics and principal component analysis [PCA] of residuals); ability to discriminate among participants' victimization and bullying behaviors (measurement precision as assessed by person separation reliability [PSR] minimum recommended value, 0.80); and targeting of items to participants' victimization and bullying.

RESULTS: Response categories were misused for both parts I and II, which required repair before further analysis. Measurement precision was inadequate for both parts (PSR, 0.64 for part I and 0.19 for part II), indicating poor discriminatory ability. All items fit the Rasch model well in part I, indicating unidimensionality that was further confirmed using PCA of residuals. However, an item misfit in part II that required deletion following which the remaining items fit and PCA of residuals also supported unidimensionality. Targeting was -0.58 logits for part I, indicating that the items were matched well with the participants' victimization. By comparison, targeting was suboptimal for part II (-1.97 logits).

CONCLUSIONS: In its current state, the revised OBVQ is not a valid psychometric instrument to assess victimization and bullying among children with VI.


Language: en

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print