I am curious to know whether methods like CFA or EFA can be used to measure macro-level constructs. For instance, I have a scale that measures a construct that is theoretically at a national level, and not appropriate for individual level analyses. Would methods like CFA or EFA be appropriate in this case to assess construct validity? (and likewise reliability, concurrent validity, etc.) Or would there be other methods that I need to take into consideration?
As long as you are able to collect a large enough sample at the macro level (this may not always be possible), I don't see why EFA or CFA could not be applied. Also, multilevel CFA allows you to test factor structures at both the individual and group levels.
Both confirmatory factor analysis (CFA) and exploratory factor analysis (EFA) are methods that can be used to assess the validity of measurement scales. These methods are typically used to assess the validity of scales that measure constructs at the individual level, but they can also be applied to assess the validity of scales that measure constructs at higher levels, such as the group or organizational level.
In the case of a scale that is designed to measure a construct at the national level, CFA or EFA could be used to assess the scale's construct validity. These methods can help to determine whether the scale is measuring the construct it was intended to measure, and can also help identify any potential problems with the scale's measurement properties.
Other methods that can be used to assess the validity of measurement scales at the national level include correlations with other measures of the construct, as well as tests of criterion validity (e.g., comparing the scale's scores to external criteria such as national statistics on the construct). It's important to consider a variety of methods when assessing the validity of a scale, as different methods can provide complementary information about the scale's measurement properties.
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts