ABSTRACT
Good evaluation requires good research practice. This chapter discusses how issues of data quality, scale, definition, and purpose can negatively affect validity and reliability of program evaluation in higher education, not only because the institutions comprising higher education vary so wildly by size, funding, clientele characteristics, purposes, and offerings, but also because 'higher education' as an entity has not been as well documented, unlike K-12 education. It examines meta-evaluative and critical approaches to what the purposes of higher education are and should be, who it should serve, and how its programs should be implemented. The issues of reliability and validity are critical to the credibility and quality of any research, regardless of the setting, research question, research design, or purpose. The chapter reviews a set of issues involving the range of research designs appropriate for research in and on higher education. Internal validity refers to whether the effect observed really is caused by the innovation or treatment under consideration.
