NML Editorial: Common Method Bias

What is it about?

Reviewers for Nonprofit Management & Leadership occasionally raise the issue of common method bias, but my sense is that our reviewers do not emphasize this issue as much as reviewers do in some other fields. This is both good and bad. The good is that reviewers and editors at some journals or in some fields misunderstand or over-simplify the problem, and thereby reject manuscripts out of hand when key variables are collected from a common source. The interdisciplinary field of nonprofit and philanthropic studies seems not to have fallen into this trap. The bad, however, is that common method bias is a potentially serious issue that does not always get the attention from authors and reviewers that it should. In this editorial, I want to offer a brief primer on the issue. A real treatment of the method bias requires a full article or chapter, so this editorial serves merely as a heads-up for those people who might be painting the issue with too broad a brush, or not at all.

Featured Image

Why is it important?

Researchers appear to have three types of options for combatting common method bias. The first is to use different methods for collecting different variables (say independent variables and outcome measures). Maybe the independent variables come from respondent self reports, but the dependent variable comes from a scrape of social media. This solution is likely to satisfy critics who see the problem simply as measurement of key variables from a common method. However, this can greatly complicate the research design, and is not an option for all studies. Also, it does not necessarily solve the structure-pattern response problem inherent to many response scales. The correlation issue may be muted when variables are collected from different sources, but the variables themselves may still suffer from various method biases. A second means of combatting common method bias is statistical control. A description of these methods is beyond the scope of this short review, but psychologists and statisticians have considered a variety of methods for detecting and correcting method bias. For a review, see Podaskoff, MacKenzie, and Podaskoff's 2012 article on method bias and its remedies. When common method bias is a clear issue for a given study, serious scholars and their critics will tread the statistical path. The third option on confronting method bias is to head it off in the research design stage. Remember, the problem arises when the conditions for response on one variable (say, independent) influences or is common to the conditions for response on another (say, dependent) variable. Anything you can do, then, to reduce that commonality can be an argument for why the variables may not suffer unduly from common method bias.

Perspectives