Increasing Quantitative Literacy in Nursing: A Joint Nursing-Statistician Perspective

Strong quantitative literacy is necessary to fulfill nurses’ professional responsibilities across education levels, roles, and settings. Evidence-based practice and systems improvement are not possible if nurses do not understand the statistics employed in generating evidence. Statistics is the language of science and rigorous nursing science cannot exist without a clear understanding of statistical methods. Increasing availability and complexity of medical and public health data and a growing focus on population health necessitate increasingly sophisticated quantitative literacy in nursing practice, leadership, and science (Hayat, Kim, Schwartz, & Jiroutek, 2021; Hayat, Schmiege, & Cook, 2014). Without strong quantitative knowledge, the nursing profession will lose opportunities to lead evidence-driven, population-focused efforts for health improvement.

Serious limitations in both knowledge and application of statistics have been documented in nursing pedagogy, scholarship, and research for decades (Anthony, 1996; Gaskin & Happell, 2014; Hayat, Higgins, Schwartz, & Staggs, 2015; Hayat et al., 2021; Hayat et al., 2014). Prior work calling for greater quantitative literacy in nursing has been solely or primarily statistician-led, leaving an opportunity and responsibility for nurses to contribute. Without the voice of nursing, efforts to improve quantitative literacy within the profession will lack nursing insight and perspective. In this commentary we provide guidance for nurses’ engagement with quantitative methods and offer suggestions to increase quantitative literacy in nursing across education levels, roles, and settings.

Be Aware of What You Do and Don’t Know

For nurses, awareness of one’s level of statistical knowledge can foster more effective communication with statisticians and consumers of nursing scholarship and avoid analytic errors ground in lack of statistical knowledge. Statistics is a discipline – not a toolbox; statistics is not simply about choosing the right analytic approach, but about a start-to-finish approach to project planning, data collection, appropriate analysis (including confirming underlying statistical assumptions and conducting sensitivity analyses), and accurately and comprehensively understanding and presenting results to a range of stakeholders with varying levels of statistical knowledge. Importantly, levels of required statistical knowledge varies by nursing role – BSN-prepared clinicians focus on evidence-based practice, MSN- and DNP-prepared nurse practitioners focuses on quality improvement and research translation to systems and practice, and PhD-prepared nurse scientists focus on creation of generalizable knowledge (Hayat et al., 2014). Thus, while BSN-, MSN-, and DNP-prepared nurses may focus on quantitative literacy, PhD-prepared nurses may recognize larger gaps in their required knowledge and focus on gaining statistical expertise required to conduct high quality nursing research.

Know How to Find Statistical Help

Many nurses – particularly those working primarily in clinical practice, in a small organization, or in a setting without formal academic-practice partnerships – may be unsure how to find statistical assistance. Often, nurses are not taught how to collaborate with statisticians (Hayat et al., 2015). A first step entails deciding from whom statistical help is needed. A graduate student in statistics, MS-prepared statistician, and PhD-prepared statistician will bring different expertise, but all may be appropriate collaborators depending on need. A nurse scientist conducting research to develop new knowledge likely requires different expertise than a hospital nursing unit manager planning a quality improvement project. Statistical collaborators are often available via academic institutions (schools of nursing but also schools of public health or medicine). Nursing-statistician collaboration typically necessitates deeper partnership than simply confirming which statistical test should be used, as statisticians can help with the entire project planning process (Hayat et al., 2015). Nursing collaborators should be aware that statistical collaboration may entail cost such as hourly fees or coverage of salary/effort. Alternatively, statistical support services may be provided by a nursing school or college, such as via consulting labs, that do not require funding or provide short term consultation as needed.

When seeking a statistical collaborator, it is important to be aware that statisticians have different focuses and areas of expertise. Simply finding “a statistician” may be too broadly defined and thus an ineffective approach. For example, the statistician who helps with instrument development may not be the same statistician who helps plan a randomized controlled trial nor the same statistician who helps analyze large, multi-level data from an electronic medical record. Attention to statisticians’ areas of expertise is important to finding the right collaborator.

For PhD-prepared Nurses, Seek Challenging Learning Opportunities

Nurses should embrace challenge when learning about statistics. For PhD-prepared nurses, gaining statistical expertise through summer intensive or short courses, formal university coursework, or career development awards is often beneficial. Particularly for nurse scientists whose work entails advanced quantitative approaches, such additional training is likely a necessity. PhD-prepared nurses should also consider challenging themselves to learn analytic tools beyond menu-driven commercial software (e.g., SPSS) (Hayat et al., 2014). Code-driven and free-of-charge statistical computing tools allow for replicability, transparency, and documentation of analytic work. R is a tool that may be of particular interest, given its open structure, large and active user-driven community, and availability of numerous flexible user-provided packages. R is also a useful tool for working with spatial data, which is relevant for nurses who are interested in social or environmental determinants of health, such as neighborhood poverty or greenspace access. Thus, while learning tools beyond menu-driven software may initially feel challenging, doing so can contribute meaningfully to one’s statistical skillset.

Focus on Quantitative Literacy Rather Than Statistical Expertise

For most nurses, the goal should be quantitative literacy (Hayat et al., 2015; Hayat et al., 2014). Nurses bring important content expertise coupled with a wealth of relevant clinical experience that can bring data analytic strategies alive for a multidisciplinary audience in a presentation or manuscript. A statistician would not take one or two courses in “nursing” and expect to care for patients. Similarly, a nurse should not take one or two biostatistics courses and aim to plan, execute, and interpret one’s own data analysis. Nurses can approach statistical collaboration with respect for the content knowledge they bring as a nurse, rather than an insecurity about the statistical expertise they lack. Through effective collaboration and a focus on strong quantitative literacy, nurses can dispel negative stereotypes about nurses not being “good at” statistics.

Advance Efforts to Increase Quantitative Literacy in Nursing

Given documented gaps in nursing knowledge, strategies for increasing quantitative literacy must be considered. Quantitative and qualitative assessments of nurses’ statistical learning needs could inform translation of the robust statistics education literature into nursing training. A nursing-focused addendum to the Guidelines for Assessment and Instruction in Statistics Education could inform nursing education, as could increased guidance on nursing education statistical competencies from accrediting bodies. In addition, formalizing processes and enforcing rigorous guidelines for manuscripts’ statistical methods sections in nursing journals could increase rigor within the nursing literature (Hayat et al., 2015; Hayat et al., 2021). Further, efforts to increase the pipeline of individuals well-prepared to serve as statisticians in schools of nursing could benefit academic nursing. Faculty who hold joint appointment in nursing and statistics and have formal education and training in both fields may be optimal, and they would have both scholarly authority in nursing and statistics and the ability to communicate effectively with nursing students. Joint graduate degree programs or minors – approaches used by many other disciplines at large research universities – can increase the pipeline of statisticians prepared in this manner and well-suited to serve in schools of nursing.

Conclusions

There is a ripe opportunity for increased nursing leadership to improve quantitative literacy in nursing. An active collaboration of nursing and statistical thought leaders can chart the path forward. When armed with appropriate statistical knowledge, nurses can play a unique role in using data to promote health and prevent disease among individuals, communities, and populations.