Stop and think about the education you amassed to get to the point you are now—the late nights studying for finals; countless hours preparing for entrance exams; the papers written, edited, and revised to convey just the right message? If you’re a Talking Heads fan you probably hear David Burns asking, “How did I get here?”

In functional terms, a reference manager is a program that allows you to collect, organize, and cite reference materials (e.g., books, journal articles, websites).  It has become an invaluable tool for anyone engaged in modern academic writing.   Through my career I've used a variety of reference managers, and their functionality is an integral part of my work flow which involves reading academic material, collaborating with other authors, and writing across a variety of platforms such as LaTeX, Authorea, and Libre Office.

Practical action research is conducted to improve practice and resolve problems of practice. This involves working within complex scenarios that can be difficult to understand. Complexity can confound otherwise well designed research, creating misunderstanding and misinterpretation within the study itself, and indistinct communication about the study. The effects of complexity can be mitigated through systematic pre-study analysis that clarifies the problem, and provides a transparent decision-trail that is examinable by other researchers and stakeholders. In this article, we present a process for conducting such analysis. The presented method provides a systematic and transparent process for analyzing problems that: a) redounds to a contextually-sensitive, high-resolution articulation of the problem; and b) prioritizes deconstructed problem elements for follow-on research. This method is presented for use as a pre-study task to enhance the effectiveness of practical action research initiatives and transferability of findings.

The process of social science scholarship - research, theoretical, methodological, and conceptual work - does not happen in isolation nor by accident. Scholarship builds on the ideas and efforts of others, challenges established orthodoxies, and provides the insights and evidence for a field of study. It is the collective process by which we've developed high resolution understandings of phenomena, solutions for complex problems, and more effective ways to flourish in our world. There is nothing natural about effective scholarship. Human nature tends toward confirmation bias, affiliatory preferences, and myopia. In other words, scholarship is a process that must be diligently maintained or it will regress back to the default characteristics of human nature. For something this important, there's surprisingly little discussion about the assumptions, norms, and habits under which we operate in scholarly environments. Perhaps it has always been assumed that those engaging in scholarship were naturally enculturated as part of their technical training. Maybe this was true at one point, but over the years, there has been a noticeable decline in many of the long-established informal scientific traditions. And to be clear, I am not referring to the differences between positivist and naturalist orientations, nor between quantitative and qualitative methods of research. What I am referring to here is a decline in a basic understanding of what constitutes good scholarship. This has been supplanted with a deterministic orientation to press a preferred narrative, and the subjugation of scholarship toward that end. That, of course, is not scholarship, it's advocacy. And while advocacy certainly has it's place, the two should not be conflated.The recognition of this decline has changed the way I introduce my  graduate students to research. Rather that assuming that certain assumptions and habits will germinate and develop naturally over time, I have begun explicitly articulating them as a launching point for discussions. As with any thesis, there are statements here that thoughtful people may disagree with; and I welcome any constructive feedback a/or criticisms.Assumptions and HabitsTo claim that you understand an idea you must be able to operationalize it, provide examples of it, and articulate what should follow from it.The scientific method does not prove, it only disproves. A theory only remains tenable through the accumulation of evidence that fails to refute it.  Research can be used selectively to support most any position. Examine the full body of evidence before accepting an evidence-based argument. Demagoguery does not edify nor does it further understanding of ideas. If you disagree with something, then present a better argument.Peer review does not necessarily mean quality. Know what constitutes quality scholarship so you can distinguish the good from the not so good.Scholarship is not immune to narrative, politics, and culture. Mind the exogenous influences when evaluating a body of scholarship.Statistical significance is not the same as practical importance. Know the established benchmarks within a field before evaluating the importance of a claim. Scholarship is grounded in reason. The integrity of an idea is what matters, not the passion that surrounds it.Affiliatory biases inhibit scholarship. Guard against reflexively accepting ideas because they are from within a familiar group or reflexively dismissing them because they are extrinsic to the group.Humility is the inevitable outcome of effective scholarship. Be cautions of those who claim to hold a monopoly on the truth.

Consider what usually happens when a large and complex problem is encountered. All too often there's a paralysis by analysis and nothing meaningful gets done at all. Sometimes solutions are identified from external sources (e.g., academic literature, colleagues, case studies) and applied. Maybe something works, maybe it doesn't. Often the metrics for improvement are fuzzy, so it's hard to tell. People become frustrated, lose interest, and move on to something else.Dr. Atul Gawande confronts this phenomena in health care delivery, describing the challenges of lowering nosocomial infection rates, a complex problem with a wide variety of potential causes. The problem was eventually addressed (hand washing of all things) not through some large scale randomized control trial or picking a solution from the proverbial hat. No, it was addressed through a short-cycle, iterative study of the problem and its possible solutions.Short-cycle ConceptuallyA short-cycle improvement study can be thought of as a phased emergent analysis of a complex problem (Figure 1). That's a loaded statement, so let's unpack the meaning behind the terms. Phased indicates that it occurs over multiple recursive steps. Emergent refers to the fact that the more we engage with a problem, the more robust our schema for it becomes. And finally, analysis means the deconstruction of a multifaceted construct into constituent elements for the purpose of better understanding it. The process has two distinct but recursive phases: an exploration phase focused on understanding the problem, and a testing phase focused on implementing and testing solutions.

Correlation is a fundamental concept within statistics that, once understood, provides insight into more complex statistical models and ideas. From a conceptual standpoint, correlation summarizes the measured association between variables, meaning the extent to which one variable is affected by the other. Put another way, correlation is simply a measure of association.The term measured association carries a lot of meaning here, so let's unpack it. To calculate the correlation between variables, we first have to measure those variables. The term measured association rather than simply association is a hedge against the possibility that those measures could be inaccurate, and not truly reflect the thing we intend to measure. Science is cautious; and the terms we use reflect that caution. The word association refers to how the data points between variables trend relative to each other. If one goes up does the other go up as well? Or maybe it goes down? Or maybe the change in one does not systematically affect the the other. Of course, this means that we need multiple data points across variables to determine correlation; but more on that in a minute.Association as a concept is a singular thing, but correlation as a measurement is multiple things. There are a variety of way to calculate correlation; and each is responsive to two important data characteristics. The first characteristic is the type of data being analyzed. All data is not created equal. It comes in levels of measurement that are categorized from least to most detailed as: nominal, ordinal, interval, ratio. Nominal is often something like a discrete category (e.g., Democrat, Republican, Libertarian, Independent) and ratio is a continuous measurement where zero represents an absence of the variable (e.g., height, age). Ordinal and interval are somewhere between. The second characteristic is the trend within the data. Data comes in different types of distributions. Imagine having a list of test scores, placing them in order from lowest to highest, plotting them on a graph, and fitting a line that summarizes the trend of the data. That line may be straight (i.e., linear) or curved (non-linear). Correlation is calculated differently based on this trend within the data begin analyzed. When the data we have is at the interval or ratio level of measurement, and we expect the trend of the data to be essentially linear, correlation is measured with a statistic known as Pearson's r, or sometimes simply r. It is measured on a continuous decimal scale from -1 to 1 (Figure 1). The number we arrive upon, called a correlation coefficient, tells us the magnitude (i.e., strength) of the association between the measured variables and the direction (i.e., positive or negative) of that association.

LaTeX is a powerful free and open-source academic writing system; however, it does come with a learning curve. This curve can be especially steep when trying to incorporate bibliographic references and formatting a document to the current writing style of the field - APA 6th edition. Anyone who has submitted a manuscript for publication is familiar with the trials and tribulations of formatting and citing sources. Any tool to help with this process is a welcome one. Although LaTeX comes close to APA compatibility natively, it doesn't quite nail it. This post describes how to set-up and integrate a powerful (and free) LaTeX editor with a powerful (and free) citation manager to assist with formatting and citing. To get started, you will need to download and install two programs. Texmaker: LaTeX editorZotero: citation managerThen download two files that will run on these programs.APA 6th File: LaTeX document formatting codeBetter BibTex: program for generating cite keysThese programs and files are compatible with the major operating systems. Installation with Windows and macOS is largely a matter of point and click. Installation on Linux requires a bit more know-how of course; and it's assumed that those using a Linux distribution are familiar with the processes for installing programs and various file extensions. If you are already familiar with the terminology of LaTeX and citation managers then read on. If not, it would be helpful to first review the key terms listed at the end of this post. Zotero Set-upZotero can be used separately as a reference manager, but its real utility comes from its ability to generate a bibliographic reference file (i.e., .bib file) that contains cite keys for each of your references. These cite keys are entered into a LaTeX document to generate citations. Although Zotero has a great deal of native functionality, some set-up is required for fluid integration with LaTeX. First, you need to install Better BibTex, a program for generating those cite keys.From the Zotero toolbar, navigate to: Tools > Add-ons > Extensions.Select the gear icon in the top-right corner and select ‘Install Add-on From File…’ Navigate to the folder where you saved the Better Bibtex file and install it. At this point you will need to restart Zotero to complete the install and auto-update to the most recent version of Better Bibtex. Navigate back to Extensions and make sure Better Bibtex is now listed. You may see an option for preferences. Don't worry about setting-up the Better BibTex preferences yet; that will be done in a moment.

At some point, we've all likely been warned that correlation is not causation. It sounds reasonable so we tend to accept the assertion, but what does it really mean? And is it always true?