Validation of the environmental analysis of mobility questionnaire (EAMQ) : comparison of complex walking tasks and the EAMQ among community dwelling older adults
Forbes, Jennifer Lynne
The primary objective of the study was to address two aspects of construct validity (i.e., face and criterion validity) of the Environmental Analysis of Mobility Questionnaire (EAMQ). The EAMQ is a self-report questionnaire, which consists of items that inquire about older adults tendencies to both encounter and avoid community mobility challenges that address several dimensions of community mobility. The EAMQ was compared to selected tasks from the Walking InCHIANTI Toolkit (WIT) and with a community mobility self-efficacy questionnaire (SE). Sixty independently living, community dwelling older adults (mean ± SD; age = 74 ± 5 years) volunteered to participate. Participation included a single visit by the researcher to the home of the participant. During the visit, demographic, health information, EAMQ, SE, and the modified WIT were completed. Regarding the first hypotheses, four of the six correlations between walking speed on the modified WIT and the EAMQ-encounter score were significant (range of significant correlations was 0.169 to 0.299; p < 0.05). By contrast, all of the correlations between walking speed on the modified WIT and EAMQ-avoidance score were significant (range of significant correlations was -0.330 to -0.410; p < 0.05. Regarding the second hypotheses, a significant positive correlation was found between SE and EAMQ-encounter (r = 0.345; p< 0.01) while a significant negative correlation was found between SE and EAMQ-avoidance (r = -0.531; p < 0.01). Furthermore, SE was significantly correlated with modified WIT performances (range of significant correlations was 0.332 to 0.578; p < 0.01). The secondary and exploratory purpose of this validation study was to determine if the EAMQ and SE both individually and additively contributed to the prediction of CWT performances. Results indicated that the EAMQ, significantly predicted walking speed on all modified WIT tasks; however, the avoidance score was the only significant predictor in the model. When SE was added to the prediction model it became the dominant and significant predictor of walking speed on most modified WIT tasks. As walking task complexity increased SE accounted for more of the variability in walking speed than the EAMQ. In conclusion, the results demonstrate partial support for the validity of the EAMQ. The EAMQ-avoidance score appears to be a valid correlate of the modified WIT and could be used as one predictor of community mobility. Recommendations are made for improvements to the EAMQ and for further investigation of its validity.
DegreeMaster of Science (M.Sc.)
DepartmentCollege of Kinesiology
ProgramCollege of Kinesiology
CommitteeFaulkner, Robert A.; Dal Bello-Haas, Vanina; Binsted, Gordon