Comparing unequal sized sample groups?


New Member
Hi all,

I have conducted a survey which includes 40 likert scale questions and is targetted at two groups; adopters and non-adopters of a certain technology. My aim is to compare and contrast adopters and non-adopters to determine which variables (encompassed within the Likert scales) discriminate and contrast both groups e.g. adopters are more open to change than non-adopters, etc.

I have targeted both adopter and non-adopter sources, however I have almost saturated my sources and have quite a significant difference in response rates from both groups; 127 adopters and only 29 non-adopters.

My feeling is the low reponses rate of non-adopters is not due to source selection bias, but more of the case that non-adopters who can comment on this technology are a very difficult population to identify.

My question is, is there any form of analysis methods that will allow me to compare these groups given the significant difference in group responses i.e. 127 vs 29?

I believe I read something about 'weighted methods' that allow you compare significantly different group sizes, but I'm finding it difficult to locate such methods of analysis.

Any help gratefully received.

Part-time PhD Candidate,
The Open University, UK.