The (Research) Proposals That Work

Proposals That Work : A Guide for Planning Dissertations and Grant Proposals

I've been reading this book by Locke et al. (1999) to improve my research skill and found the following quotes useful.

Introducing a study:

Proposals...are best introduced by a short, meticulously devised statement that establishes the overall area of concern, arouses interest, and communicates information essential to the reader's comprehension of what follows (page 9).

...avoids both tedious length and the shock of technical detail or abstruse argument... (page 9)

Providing a rationale:

...it often is helpful to diagram factors and relationships that support your formulation of the problem (page 10).

A sound rationale...convinces the reader that you are rasing the right question -- and that the answer is worth finding (page 10).

...justifying the proposed study should be limited to the basic matters of defining what is to be studied and why it is worth so doing (page 11).

Formulating questions:

...the proposed study should be directed toward outcomes that are foreshadowed by the literature or pilot work, rather than toward a scanning of potentially interesting findings... (page 12)

The question form is appropriate when the research is exploratory. However, we should indicate by the specificity of questions that the problem has been subject to thorough analysis. The hypothesis form is employed when the state of existing knowledge and theory permits formulation of reasonable predictions about the relationship of variables (page 12).

A clear (research) question is readily transformed into a hypothesis by casting it in the form of a declarative statement that can be tested so as to show it to be either true or false (page 14).

...a research question never permits the investigator to say more than "Here is how the world looked when I observed it." In contrast, hypotheses permits the investigator to say, "Based on my particular explanation of how the world works, this is what I expect to observe, and behond -- that is exactly how it looked! For that reason my explanation of how the world works must be given credibility (page 14).

Credibility: Note the current debate about the value of hypotheses and their associted statistical significance testing. It's been argued that statistical significance testing has certain technical limitations. We should discuss with advisors and committee members for a concensus (page 15).

Employ "directional hypotheses" only when pilot data are available that clearly indicate a direction, or when the theory from which the hypotheses were drawn is sufficiently robust, otherwise employing "null hypotheses" is the better choice (page 15).

For everyday's work, let's keep in mind "What is my research question?"


Keywords: research-methodology, quotes