NotesFAQContact Us
Search Tips
ERIC Number: ED353304
Record Type: Non-Journal
Publication Date: 1992-Nov
Pages: 31
Abstractor: N/A
Reference Count: N/A
Assessing Construct Validity in Community Evaluations: A Multitrait-Multimethod Approach.
Babcock, Judith L.; And Others
This study used multiple methods to assess basic community needs and attributes of community atmosphere (cohesion, religious involvement, and recreational activities) in two psychometric studies. Part 1 revised self-report community assessment measures, developed multi-item scales for each construct, and tested reliabilities and factor structures of the 22-item Community Needs Index (CNI) and the 40-item Community Environment Scale (CES). The CNI and CES were given to 27 members of civic organizations in 2 rural communities. Part 2 used a non-verbal assessment method to assess the same constructs as the self-report measures and added photo observation and photo-assisted interviews. Three measures were used across nine community dimensions (cohesion-concern, rural uniqueness, community pride, intellectual/cultural activities, religious involvement, conflict, outgoingness/involvement, local resources, and openness/expressiveness) to produce a multitrait-multimethod (MTMM) approach to community assessment. The MTMM matrix included data from the CNI, CES, photo observations, and photo-interviews; and used six CES scales. The construct validity of community assessment measures should not be taken for granted. Different methods do not assess the same constructs. Method variance, not substantive variance, may play a large role in the reliable variance captured by a specific trait-method-unit. Overlapping measures can be combined to enhance reliability and reduce irrelevant method effects. One table and three figures are included. (RLC)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A