Skip to the content

Development and validation of a psychometric scale for assessing PA chest image quality : A pilot study

Mriaty, H, England, A, Akhtar, I, Aslam, A, De Lange, R, Momoniat, H, Nicoulaz, S, Ribeiro, A, Mazhir, S and Hogg, P 2014, 'Development and validation of a psychometric scale for assessing PA chest image quality : A pilot study' , Radiography, 20 (4) , pp. 312-317.

[img] PDF
Restricted to Repository staff only

Download (549kB) | Request a copy


Purpose: To develop and validate a psychometric scale for assessing image quality perception for chest X-ray images. Methods: Bandura's theory was used to guide scale development. A review of the literature was undertaken to identify items/factors which could be used to evaluate image quality using a perceptual approach. A draft scale was then created (22 items) and presented to a focus group (student and qualified radiographers). Within the focus group the draft scale was discussed and modified. A series of seven postero-anterior chest images were generated using a phantom with a range of image qualities. Image quality perception was confirmed for the seven images using signal-to-noise ratio (SNR 17.2–36.5). Participants (student and qualified radiographers and radiology trainees) were then invited to independently score each of the seven images using the draft image quality perception scale. Cronbach alpha was used to test interval reliability. Results: Fifty three participants used the scale to grade image quality perception on each of the seven images. Aggregated mean scale score increased with increasing SNR from 42.1 to 87.7 (r = 0.98, P < 0.001). For each of the 22 individual scale items there was clear differentiation of low, mid and high quality images. A Cronbach alpha coefficient of >0.7 was obtained across each of the seven images. Conclusion: This study represents the first development of a chest image quality perception scale based on Bandura's theory. There was excellent correlation between the image quality perception scores derived using the scale and the SNR. Further research will involve a more detailed item and factor analysis.

Item Type: Article
Themes: Health and Wellbeing
Schools: Schools > School of Health Sciences > Centre for Health Sciences Research
Schools > School of Health Sciences
Journal or Publication Title: Radiography
Publisher: Elsevier
Refereed: No
ISSN: 1078-8174
Funders: Non funded research
Depositing User: P Hogg
Date Deposited: 23 May 2014 16:20
Last Modified: 30 Nov 2015 23:42

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)


Downloads per month over past year