Web-based measurement: Effect of completing single or multiple items per webpage

• Published in: Computers in human behavior Vol. 25; no. 2; pp. 393 - 401
• Main Authors: Thorndike, Frances P., Carlbring, Per, Smyth, Frederick L., Magee, Joshua C., Gonder-Frederick, Linda, Ost, Lars-Göran, Ritterband, Lee M.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01-03-2009; Elsevier
• Subjects: Adult and adolescent clinical studies; Anxiety disorders. Neuroses; Assessment; Biological and medical sciences; Depression; Fundamental and applied biological sciences. Psychology; Measurement; Medical sciences; Mood disorders; Online; Phobia; Psychology; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Psychometrics; Psychometrics. Diagnostic aid systems; Psychometrics. Statistics. Methodology; Psychopathology. Psychiatry; Psykologi; SAMHÄLLSVETENSKAP; SOCIAL SCIENCES; Socialvetenskap; Techniques and methods; Web; Measurement; Online; Web; Assessment; Mood disorder; Human; Evaluation; Help seeking; Panic; On line; Questionnaire; Social phobia; Anxiety disorder; Psychometrics; Depression; Internet; Information layout; Web site
• ISSN: 0747-5632; 1873-7692; 1873-7692
• DOI: 10.1016/j.chb.2008.05.006

The current study was conducted to determine whether participants respond differently to online questionnaires presenting all items on a single webpage versus questionnaires presenting only one item per page, and whether participants prefer one format over the other. Of participants seeking self-help treatment on the Internet (for depression, social phobia, or panic disorder), 710 completed four questionnaires (Beck Depression Inventory, Beck Anxiety Inventory, Quality of Life Index, Montgomery-Åsberg Depression Rating Scale) on the Internet on two occasions. The questionnaires were either presented with one questionnaire on one webpage (e.g., BDI on one webpage) or on multiple webpages (e.g., BDI on 21 webpages with one item each). Results suggest that the four web questionnaires measure the same construct across diagnostic group (depression, social phobia, panic), presentation type (single versus multiple items per page), and order of presentation (which format first). Within each diagnostic group, factor means for all questionnaires were equivalent across presentation method and time. Furthermore, factor means varied as expected across samples (e.g., depressed group scored higher on BDI), providing evidence of construct validity. The majority of participants in each diagnostic group preferred the single item per page format, even though this format required more time.