Parsimony

Quantitative Research

In the literature, quantitative researchers frequently address parsimony in the data or analysis sections. It is not uncommon to see, for example, “in the interest of parsimony” or “to maximize parsimony” as scholars justify statistical procedures and/or the development of a theoretical construct. Common strategies for increasing parsimony cited in the literature include the following:

• Moving up a level of abstraction.
• Combining data or variables to create a composite measure.
• Deleting items from a scale that are deemed irrelevant or statistically insignificant to the phenomenon under investigation.
• Selecting measures and excluding others based on level of measured parsimoniousness.
• Dropping a term from an explanatory model, such as a linear regression equation, that is deemed unnecessary.
• If structural modeling is being used to test the model or theoretical construct, calculating the ratio of chi-square to degrees of freedom (CMIN/DF) can be used as a measure of parsimony.

Qualitative Research

When it comes to parsimony in the literature, the focus is on theory derived from quantitative research and the “hard sciences.” However, the communication discipline uses qualitative research methods to explain the communication dynamics associated with a given phenomenon. The data collected offer a more personalized and in-depth explanation of an observed phenomenon. For example, a communication scholar studying the pregnancy-related experiences of low-income mothers in the inner city may conduct in-depth interviews or focus groups to learn more about the experiences and establish common themes to describe the collective. However, there is less focus on a formulaic theory or expression that unequivocally predicts the experience for all members of the populations; the focus is more on the individual experience in an attempt to find commonality. As such, parsimony is less of a focus in qualitative research. However, the lessons of simplicity and necessity, the essence of parsimony, can also be applied to qualitative methods. For example, limiting the number of identified themes or collapsing categories (i.e., moving up a level of abstraction) to only those necessary to describe the communication phenomenon avoids overly complicated explanations where relationships between themes are not fully explicated. Likewise, a sort of linguistic parsimony or simplicity of word choice (i.e., avoiding being overly verbose) offers a word of caution for qualitative researchers tempted to include quote after quote in an attempt to accurately reflect the experiences of their participants. Selective use of quoted content avoids information overload and encourages researchers to only include quotes or examples that exemplify a given theme or category.

...