Evaluating Random Error in Clinician-Administered Surveys: Theoretical Considerations and Clinical Applications of Interobserver Reliability and Agreement

Abstract

This study demonstrates that the methodology used to evaluate survey score concordance can influence the statistical results obtained and thus affect clinical interpretations.

The Influence of Destination Image on Tourist Loyalty and Intention to Visit: Testing a Multiple Mediation Approach

“…We used several fit signs, which incorporated a TLI, CFI, SRMR, IFI, RMSEA, and a chi-square test (χ 2 ) [84] to validate the model fit. As per Bennett et al [85], the values of χ 2 /df must be lower than 3 for a model fit [86], and the values of "CFI, IFI, and TLI should be greater than 0.90" [87]. SRMR and RMSEA values should not exceed 0.08 [88].…”

The influence of content and non-content cues of tourism information quality on the creation of destination image in social media: A study of Khyber Pakhtunkhwa, Pakistan

“…SRMR and RMSEA values not exceeding 0.08 (McNeish, 2018). As per (Bennett et al, 2017), for good model fit, x 2 /df must be less than 3 (Kanwel et al, 2019).The fit measurements are shown in table 2. The values of this research are higher than the recommended standards (McNeish, 2018).…”

Evaluating Hearing Aid Management: Development of the Hearing Aid Skills and Knowledge Inventory (HASKI)

“…A dichotomous scale was chosen because a pilot study (Bennett et al, 2017a) suggested that a client requires additional support regardless of whether they show complete or partial incompetence. Additionally, a recent tutorial on inter-observer concordance suggested that the low inter-rater reliability observed with currently available hearing aid skills survey is likely due to: 1) having a greater number of possible responses to an item; 2) use of survey items that are unclear and require the administrator to provide explanation throughout testing and 3) where response items are not defined and require the administrator to make their own interpretation of whether the participants response is sufficient (Bennett, Taljaard, Olaithe, Brennan-Jones & Eikelboom, 2017b). For example, when asking a participant to demonstrate how they clean their hearing aid, administrators may have differing concepts of what specific tasks are required to fulfil this request.…”