descriptive analysis nursing research

Research 101: Descriptive statistics

Use these tools to analyze data vital to practice-improvement projects..

By Brian Conner, PhD, RN, CNE and Emily Johnson, PhD

• Nurses at every level should be able to understand and apply basic statistical analyses related to performance improvement projects.
• Measures of central tendency (such as mean) and variability (such as standard deviation) are fairly common and easy to use.

How many times have you said (or heard), “Statistics are too complicated”? A significant percentage of graduate students and nurses in clinical practice report feeling anxious when working with statistics. And although some statistical analysis is pretty complicated, you don’t need a doctoral degree to understand and use descriptive statistics.

What are descriptive statistics?

Sometimes, descriptive statistics are the only analyses completed in a research or evidence-based practice study; however, they don’t typically help us reach conclusions about hypotheses. Instead, they’re used as preliminary data, which can provide the foundation for future research by defining initial problems or identifying essential analyses in more complex investigations.

Common descriptive statistics

The most common types of descriptive statistics are the measures of central tendency (mean, median, and mode) that are used in most levels of math, research, evidence-based practice, and quality improvement. These measures describe the central portion of frequency distribution for a data set.

The most familiar of these is the mean , or average, which most people use and understand. It’s calculated by adding the sum of values in the data and dividing by the total number of observations.

The median is a number found at the exact middle of a set of data. If there are two numbers at the middle of the data set (which occurs when there is an even number of data points), these two numbers are averaged to identify the median. It’s typically used to describe a data set that has extreme outliers (very low or very high numbers, distant from the majority of data points), in which case the mean will not accurately represent the data. (See What to do with outliers .) To calculate a mean or median, data must be quantitative/continuous (have an infinite number of possibilities).

What to do with outliers

The mode represents the most frequently occurring number or item in a data set. Some data sets have more than one mode, making them bimodal (two modes) or multimodal (more than two modes). The mode can be calculated with data that are quantitative/continuous or qualitative/categorical (have a finite number of categories or groups, such as sex, race, or education level). The mode is the only measure of central tendency that can be analyzed with qualitative/categorical data.

Less common descriptive statistics

Measures of variability or dispersion are less common descriptive statistics, but they’re still important because they describe the spread of values across a data set. Although the central tendency of data is vital, the range of values (the difference between the maximum and minimum values in the data) also may be important to note. The range not only sets boundaries for your data set and indicates the spread, but it also can identify errors in the data. For example, if you have a data set with a diastolic blood pressure range of 230 (highest diastolic value) to 25 (lowest diastolic value) = 205 (range), an error probably exists in your data because the values of 230 and 25 aren’t valid blood pressure measures in most studies. Other measures of variability include standard deviation, variance, and quartiles. (See Other variability measures .)

Other variability measures

Practical application of descriptive statistics.

To put all of this information into perspective, here’s an example of how these measures can be used in a clinical setting.

A rural primary care clinic has a high percentage of patients with diabetes whose glycated hemo­globin (HbA1c) levels are greater than 7% (uncontrolled HbA1c) and body mass index (BMI) is over 30. The clinic implements a 9-month quality-improvement initiative to lower these numbers. The initiative includes a wellness education program focused on exercise, healthy eating, and understanding the importance of regular blood glucose monitoring . Before implementing the program, the clinic collects 3 months of aggregate data (3, 6, and 9 months before the intervention) for all patients with diabetes in the clinic, including HbA1c levels, BMIs, and patients with uncontrolled HbA1c. Gender and age also are collected. The clinic then collects the same data 3, 6, and 9 months after implementation of the program. (See Snapshot of aggregate data .) Because of the different types of data collected, different measures of central tendency and variability can help describe outcomes. (See Statistical analysis examples .)

Snapshot of aggregate data

Statistical analysis examples

Implications for practice

Nurses are increasingly asked to participate and lead evidence-based practice and quality-improvement projects. Many healthcare organizations, including those aspiring to or holding Magnet® recognition from the American Nurses Credentialing Center, require that nurses take part in these activities to achieve higher levels of professional development within clinical ladder programs. Nurses can and should learn how to use descriptive statistics to analyze and depict vital data related to practice-improvement projects.

Brian Conner is adjunct faculty at the School of Nursing and Health Sciences for Simmons College in Boston, Massachusetts. Emily Johnson is an assistant professor at the Medical University of South Carolina College of Nursing in Charleston.

Selected references

American Nurses Credentialing Center (ANCC). Magnet Recognition Program® Overview . 2016.

Heavey E. Statistics for Nursing: A Practical Approach . 2nd ed. Burlington, MA: Jones and Bartlett Learning; 2015.

Thabane L, Akhtar-Danesh N. Guidelines for reporting descriptive statistics in health research. Nurse Res . 2008;15(2):72-81.

Zhang Y, Shang L, Wang R, et al. Attitudes toward statistics in medical postgraduates: Measuring, evaluating and monitoring. BMC Med Educ . 2012;12:117.

1 Comment .

This is a great data set and I would like to see if I can have copy right to use it in my « statistical thinking for Nursing » course here at Castleton Unvesity to demonstrate Bootsrapjng technique as Inferential statistics. Please let me know. Thank You Dr. Rajia

Comments are closed.

NurseLine Newsletter

• First Name *
• Last Name *
• Hidden Referrer

*By submitting your e-mail, you are opting in to receiving information from Healthcom Media and Affiliates. The details, including your email address/mobile number, may be used to keep you informed about future products and services.

Test Your Knowledge

Recent posts.

Interpreting statistical significance in nursing research

Introduction to qualitative nursing research

Navigating statistics for successful project implementation

Nurse research and the institutional review board

Research 101: Forest plots

Understanding confidence intervals helps you make better clinical decisions

Differentiating statistical significance and clinical significance

Differentiating research, evidence-based practice, and quality improvement

Are you confident about confidence intervals?

Making sense of statistical power

The P value: What it really means

• Counseling Programs Overview
• Clinical Mental Health Counseling
• Professional School Counseling
• Nursing Programs Overview
• Master of Science in Nursing
• Doctor of Nursing Practice
• Post-Master’s Certificate
• Bachelor of Science in Speech Language and Hearing Sciences
• Nursing Resources
• Counseling Resources

What's the difference between descriptive and inferential statistics?

January 16, 2024

View all blog posts under Articles | View all blog posts under Nursing Resources

However, the use of data goes well beyond storing electronic health records (EHRs). Increasingly, insights are driving provider performance, aligning performance with value-based reimbursement models, streamlining health care system operations, and guiding care delivery improvements. As a result, DNP-prepared nurses are now more likely to have some proficiency in statistics and are expected to understand the intersection of statistical analysis and health care.

For nurses to succeed in leveraging these types of insights, it’s crucial to understand the difference between descriptive statistics vs. inferential statistics and how to use both techniques to solve real-world problems. According to the American Nurses Association (ANA), nurses at every level should be able to understand and apply basic statistical analyses related to performance improvement projects.

In Bradley University’s online DNP program , students study the principles and procedures of statistical interpretation. Here’s what nursing professionals need to know about descriptive and inferential statistics, and how these types of statistics are used in health care settings.

What Are Descriptive Statistics?

In essence, descriptive statistics are used to report or describe the features or characteristics of data. They summarize a particular numerical data set,or multiple sets, and deliver quantitative insights about that data through numerical or graphical representation.

Descriptive statistics only reflect the data to which they are applied. A descriptive statistic can be:

• A measure of central tendency, like mean, median, or mode: These are used to identify an average or center point among a data set
• A measure of dispersion or variability, like variance, standard deviation, skewness, or range: These reflect the spread of the data points
• A measure of distribution, like the quantity or percentage of a particular outcome: These express the frequency of that outcome among a data set

Virtually any quantitative data can be analyzed using descriptive statistics, like the results from a clinical trial related to the side effects of a particular medication.

Descriptive statistics expressing a measure of central tendency might show the mean age of people who tried the medication was 37. Additionally, as a measure of distribution, descriptive statistics could show 25% of the group experienced mild side effects, while 2% felt moderate to severe side effects and 73% felt no side effects.

The raw data can be represented as statistics and graphs, using visualizations like pie charts, line graphs, tables, and other representations summarizing the data gathered about a given population.

What Are Inferential Statistics?

Inferential statistics are used to make conclusions, or inferences, based on the available data from a smaller sample population. This is often done by analyzing a random sampling from a much broader data set, like a larger population. Conclusions drawn from this sample are applied across the entire population.

The relevance and quality of the sample population are essential in ensuring the inference made is reliable. This is true whether the population is a group of people, geographic areas, health care facilities, or something else entirely. A representative sample must be large enough to result in statistically significant findings, but not so large it’s impossible to analyze.

Inferential statistics techniques include:

• Hypothesis tests, or tests of significance: These involve confirming whether certain results are significant and not simply by chance
• Correlation analysis: This helps determine the relationship or correlation between variables
• Logistic or linear regression analysis: These methods enable inferring and predicting causality and other relationships between variables
• Confidence intervals: These help identify the probability an estimated outcome will occur

As an example, inferential statistics may be used in research about instances of comorbidities. Instead of canvassing vast health care records in their entirety, researchers can analyze a sample set of patients with shared attributes — like those with more than two chronic conditions — and extrapolate results across the larger population from which the sample was taken.

Studying a random sample of patients within this population can reveal correlations, probabilities, and other relationships present in the patient data. These findings may help inform provider initiatives or policymaking to improve care for patients across the broader population.

What’s the Difference Between Descriptive Statistics vs. Inferential Statistics?

The key difference between descriptive and inferential statistics is descriptive statistics aren’t used to make an inference about a broader population, whereas inferential statistics are used for this purpose. Rather than being used to report on the data set itself, inferential statistics are used to generate insights across vast data sets that would be difficult or impossible to analyze.

Essentially, descriptive statistics state facts and proven outcomes from a population, whereas inferential statistics analyze samplings to make predictions about larger populations.

In the example of a clinical drug trial, the percentage breakdown of side effect frequency and the mean age represents statistical measures of central tendency and normal distribution within that data set.

However, inferential statistics methods could be applied to draw conclusions about how such side effects occur among patients taking this medication. The resulting inferential statistics can help doctors and patients understand the likelihood of experiencing a negative side effect, based on how many members of the sample population experienced it.

Since descriptive statistics focus on the characteristics of a data set, the certainty level is very high. Outliers and other factors may be excluded from the overall findings to ensure greater accuracy, but calculations are often much less complex and can result in solid conclusions.

However, inferential statistics are designed to test for a dependent variable — namely, the population parameter or outcome being studied — and may involve several variables. The calculations are more advanced, but the results are less certain. There will be a margin of error as well. After all, inferential statistics are more like highly educated guesses than assertions. A sampling error may skew the findings, although a variety of statistical methods can be applied to minimize problematic results.

When Should You Use Descriptive Statistics vs. Inferential Statistics in Nursing?

For example, nurse executives who oversee budgeting and other financial responsibilities will likely need familiarity with descriptive statistics and their use in accounting. Descriptive statistics can also come into play for professionals like family nurse practitioners or emergency room nurse managers who must know how to calculate variance in a patient’s blood pressure or blood sugar. Moreover, in a family clinic, nurses might analyze the body mass index (BMI) of patients at any age.

But descriptive statistics only make up part of the picture, according to the journal American Nurse. Sometimes, descriptive statistics are the only analyses completed in a research or evidence-based practice study; however, they don’t typically help us reach conclusions about hypotheses. Instead, they’re used as preliminary data, which can provide the foundation for future research by defining initial problems or identifying essential analyses in more complex investigations.

Following up with inferential statistics can be an important step toward improving care delivery, safety, and patient experiences across wider populations. Since it’s virtually impossible to survey all patients who share certain characteristics, Inferential statistics are crucial in forming predictions or theories about a larger group of patients. The sample data can indicate broader trends across the entire population.

Such statistics have clear use regarding the rise of population health. For instance, examining the health outcomes and other data of patient populations like minority groups, rural patients, or seniors can help nurse practitioners develop better initiatives to improve care delivery, patient safety, and other facets of the patient experience. Looking at how a sample set of rural patients responded to telehealth-based care may indicate it’s worth investing in such technology to increase telehealth service access. Techniques like hypothesis testing and confidence intervals can reveal whether certain inferences will hold up when applied across a larger population.

How Can You Learn About Statistics for Nursing?

Aspiring leaders in the nursing profession must be confident in using statistical analysis to inform empirical research and therefore guide the creation and application of evidence-based practice methods. At Bradley University , the online Doctor of Nursing Practice program prepares students to leverage these techniques in health care settings. This is true of both DNP tracks at Bradley, namely:

• The DNP-FNP track , for those who enroll with a Bachelor of Science in Nursing (BSN) and aspire to become family nurse practitioners (FNPs)
• The DNP-Leadership track , for those who enroll with a Master of Science in Nursing (MSN) and aspire to health care leadership roles

The curricula of both the DNP-FNP and DNP-Leadership programs include courses intended to impart key statistical knowledge and data analysis skills to be used in a nursing career, such as:

• Research Design and Statistical Methods
• Methods in Evidence Based Practice
• Advanced Health Informatics

Research Design and Statistical Methods introduces an examination of research study design/methodology, application, and interpretation of descriptive and inferential statistical methods appropriate for critical appraisal of evidence. Use of analytic software for data management and preliminary analysis prepares students to assess quantitative and qualitative data, understand research methodology, and critically evaluate research findings.

Methods in Evidence Based Practice introduces students to theories related to Research Utilization (RU) and Evidence-based Practice (EBP) and provides opportunities to explore issues and refine questions related to quality and cost-effective healthcare delivery for the best client outcomes. Methods to collect evidence, plan changes for the transformation of practice, and evaluate quality improvement methods will be discussed.

Finally, the Advanced Health Informatics course examines the current trends in health informatics and data analytic methods. It provides opportunities for the advanced practice nurse (APN) to apply theoretical concepts of informatics to individual and aggregate level health information. Emphasis is placed on the APN’s leadership role in the use of health information to improve health care delivery and outcomes.

Gain Statistical Expertise in Bradley’s Online DNP Program

Bradley’s online DNP program offers nursing students a flexible learning environment that can work around their existing personal and professional needs.

The DNP-FNP track is offered 100% online with no campus residency requirements. It involves completing 10 semesters and 1,000 clinical hours, which takes full-time students approximately 3.3 years to complete.

The DNP-Leadership track is also offered 100% online, without any campus residency requirements. This program involves finishing eight semesters and 1,000 clinical hours, taking students 2-2.7 years to complete if they study full time.

Interested in learning more about where an online DNP could take your nursing career? Visit our online DNP program page and contact an enrollment advisor today for more information.

Recommended Readings:

Online Doctor of Nursing Practice

Principles of Nursing Leadership: Jobs and Trends

Career Profile: Nursing Professor Salaries, Skills, and Responsibilities

American Nurse — Research 101: Descriptive Statistics

Indeed — Descriptive vs Inferential Statistics

ThoughtCo — The Difference Between Descriptive and Inferential Statistics

Investopedia — Descriptive Statistics

My Market Research Methods — Descriptive vs Inferential Statistics: What’s the Difference?

Researchgate — Interpretation and Use of Statistics in Nursing Research

Learn more about Bradley’s Online Degree Programs.

Understanding Descriptive Research Designs and Methods

Affiliation.

• 1 Author Affiliation: Senior Nurse Scientist and Clinical Nurse Specialist, Office of Nursing Research & Innovation, Nursing Institute, Cleveland Clinic, Ohio.
• PMID: 31789957
• DOI: 10.1097/NUR.0000000000000493
• Nurse Clinicians / psychology*
• Nursing Research / methods*
• Research Design*

An official website of the United States government

Official websites use .gov A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

• Publications
• Account settings
• Advanced Search
• Journal List

Qualitative Descriptive Methods in Health Science Research

Karen jiggins colorafi , phd, mba, rn, bronwynne evans , phd, rn, fngna, anef, faan.

• Author information
• Article notes
• Copyright and License information

Corresponding Author: Karen Jiggins Colorafi, PhD, MBA, RN, College of Nursing & Health Innovation, Arizona State University, 550N. 3rd Street, Phoenix, AZ 85004, USA. [email protected]

Issue date 2016 Jul.

Reprints and permission: sagepub.com/journalsPermissions.nav

The purpose of this methodology paper is to describe an approach to qualitative design known as qualitative descriptive that is well suited to junior health sciences researchers because it can be used with a variety of theoretical approaches, sampling techniques, and data collection strategies.

Background:

It is often difficult for junior qualitative researchers to pull together the tools and resources they need to embark on a high-quality qualitative research study and to manage the volumes of data they collect during qualitative studies. This paper seeks to pull together much needed resources and provide an overview of methods.

A step-by-step guide to planning a qualitative descriptive study and analyzing the data is provided, utilizing exemplars from the authors’ research.

This paper presents steps to conducting a qualitative descriptive study under the following headings: describing the qualitative descriptive approach, designing a qualitative descriptive study, steps to data analysis, and ensuring rigor of findings.

Conclusions:

The qualitative descriptive approach results in a summary in everyday, factual language that facilitates understanding of a selected phenomenon across disciplines of health science researchers.

Keywords: qualitative descriptive, qualitative methodology, rigor, qualitative design, qualitative analysis

There is an explosion in qualitative methodologies among health science researchers because social problems lend themselves toward thoughtful exploration, such as when issues of interest are complex, have variables or concepts that are not easily measured, or involve listening to populations who have traditionally been silenced ( Creswell, 2013 ). Creswell (2013 , p. 48) suggests qualitative research is preferred when health science researchers seek to (a) share individual stories, (b) write in a literary, flexible style, (c) understand the context or setting of issues, (d) explain mechanisms or linkages in causal theories, (e) develop theories, and (f) when traditional quantitative statistical analyses do not fit the problem at hand. Typically, qualitative textbooks present learners with five approaches for qualitative inquiry: narrative, phenomenological, grounded theory, case study, and ethnography. Yet eminent researcher Margarete Sandelowski argues that in “the now vast qualitative methods literature, there is no comprehensive description of qualitative description as a distinctive method of equal standing with other qualitative methods, although it is one of the most frequently employed methodological approaches in the practice disciplines” ( Sandelowski, 2000 ). Qualitative description is especially amenable to health environments research because it provides factual responses to questions about how people feel about a particular space, what reasons they have for using features of the space, who is using particular services or functions of a space, and the factors that facilitate or hinder use.

The purpose of this methodology article is to define and outline qualitative description for health science researchers, providing a starter guide containing important primary sources for those who wish to become better acquainted with this methodological approach.

Describing the Qualitative Descriptive Approach

In two seminal articles, Sandelowski promotes the mainstream use of qualitative description ( Sandelowski, 2000 , 2010 ) as a well-developed but unacknowledged method which provides a “comprehensive summary of an event in the every day terms of those events” ( Sandelowski, 2000 , p. 336). Such studies are characterized by lower levels of interpretation than are high-inference qualitative approaches such as phenomenology or grounded theory and require a less “conceptual or otherwise highly abstract rendering of data” ( Sandelowski, 2000 , p. 335). Researchers using qualitative description “stay closer to their data and to the surface of words and events” ( Sandelowski, 2000 , p. 336) than many other methodological approaches. Qualitative descriptive studies focus on low-inference description, which increases the likelihood of agreement among multiple researchers. The difference between high and low inference approaches is not one of rigor but refers to the amount of logical reasoning required to move from a data-based premise to a conclusion. Researchers who use qualitative description may choose to use the lens of an associated interpretive theory or conceptual framework to guide their studies, but they are prepared to alter that framework as necessary during the course of the study ( Sandelowski, 2010 ). These theories and frameworks serve as conceptual hooks upon which hang study procedures, analysis, and re-presentation. Findings are presented in straightforward language that clearly describes the phenomena of interest.

Other cardinal features of the qualitative descriptive approach include (a) a broad range of choices for theoretical or philosophical orientations, (b) the use of virtually any purposive sampling technique (e.g., maximum variation, homogenous, typical case, criterion), (c) the use of observations, document review, or minimally to moderately structured interview or focus group questions, (d) content analysis and descriptive statistical analysis as data analysis techniques, and (e) the provision of a descriptive summary of the informational contents of the data organized in a way that best fits the data ( Neergaard, Olesen, Andersen, & Sondergaard, 2009 ; Sandelowski, 2000 , 2001 , 2010 ).

Designing a Qualitative Descriptive Study

Methodology.

Unlike traditional qualitative methodologies such as grounded theory, which are built upon a particular, prescribed constellation of procedures and techniques, qualitative description is grounded in the general principles of naturalistic inquiry. Lincoln and Guba suggest that naturalistic inquiry deals with the concept of truth, whereby truth is “a systematic set of beliefs, together with their accompanying methods” ( Lincoln & Guba, 1985 , p. 16). Using an often eclectic compilation of sampling, data collection, and data analysis techniques, the researcher studies something in its natural state and does not attempt to manipulate or interfere with the ordinary unfolding of events. Taken together, these practices lead to “true understanding” or “ultimate truth.” Table 1 describes design elements in two exemplar qualitative descriptive studies and serves as guide to the following discussion.

Example of Study Design Elements for Two Studies.

Adapted from Jiggins Colorafi (2015) .

Adapted from Evans, Belyea, Coon, and Ume (2012) ; Evans, Belyea, and Ume (2011)

Theoretical Framework

Theoretical frameworks serve as organizing structures for research design: sampling, data collection, analysis, and interpretation, including coding schemes, and formatting hypothesis for further testing ( Evans, Coon, & Ume, 2011 ; Miles, Huberman, & Saldana, 2014 ; Sandelowski, 2010 ). Such frameworks affect the way in which data are ultimately viewed; qualitative description supports and allows for the use of virtually any theory ( Sandelowski, 2010 ). Creswell’s chapter on “Philosophical Assumptions and Interpretative Frameworks” (2013) is a useful place to gain understanding about how to embed a theory into a study.

Sampling choices place a boundary around the conclusions you can draw from your qualitative study and influence the confidence you and others place in them ( Miles et al., 2014 ). A hallmark of the qualitative descriptive approach is the acceptability of virtually any sampling technique (e.g., maximum variation where you aim to collect as many different cases as possible or homogenous whereby participants are mostly the same). See Miles, Huberman, and Saldana’s (2014 , p. 30) “Bounding the Collection of Data” discussion to select an appropriate and congruent purposive sampling strategy for your qualitative study.

Data Collection

In qualitative descriptive studies, data collection attempts to discover “the who, what and where of events” or experiences ( Sandelowski, 2000 , p.339). This includes, but is not limited to focus groups, individual interviews, observation, and the examination of documents or artifacts.

Data Analysis

Content analysis refers to a technique commonly used in qualitative research to analyze words or phrases in text documents. Hsieh and Shannon (2005) present three types of content analysis, any of which could be used in a qualitative descriptive study. Conventional content analysis is used in studies that aim to describe a phenomenon where exiting research and theory are limited. Data are collected from open-ended questions, read word for word, and then coded. Notes are made and codes are categorized. Directed content analysis is used in studies where existing theory or research exists: it can be used to further describe phenomena that are incomplete or would benefit from further description. Initial codes are created from theory or research and applied to data and unlabeled portions of text are given new codes. Summative content analysis is used to quantify and interpret words in context, exploring their usage. Data sources are typically seminal texts or electronic word searches.

Quantitative data can be included in qualitative descriptive studies if they aim to more adequately or fully describe the participants or phenomenon of interest. Counting is conceptualized as a “means to and end, not the end itself” by Sandelowski (2000 , p. 338) who emphasizes that careful descriptive statistical analysis is an effort to understand the content of data, not simply the means and frequencies, and results in a highly nuanced description of the patterns or regularities of the phenomenon of interest ( Sandelowski, 2000 , 2010 ). The use of validated measures can assist with generating dependable and meaningful findings, especially when the instrument (e.g., survey, questionnaire, or list of questions) used in your study has been used in others, helping to build theory, improve predictions, or make recommendations ( Miles et al., 2014 ).

Data Re-Presentation

In clear and simple terms, the “expected outcome of qualitative descriptive studies is a straight forward descriptive summary of the informational contents of data organized in a way that best fits the data” ( Sandelowski, 2000 , p. 339). Data re-presentation techniques allow for tremendous creativity and variation among researchers and studies. Several good resources are provided to spur imagination ( Miles et al., 2014 ; Munhall & Chenail, 2008 ; Wolcott, 2009 ).

Steps to Data Analysis

It is often difficult for junior health science researchers to know what to do with the volumes of data collected during a qualitative study and formal course work in traditional qualitative methods courses are typically sparse regarding the specifics of data management. It is for those reasons that this section of our article will provide a detailed description of the data analysis techniques used in qualitative descriptive methodology. The following steps are case examples of a study undertaken by one author (K.J.C.) after completing a data management course offered by another author (B.E.). Examples are offered from the two studies noted in Table 1 . It is offered in list format for general readability, but the qualitative researcher should recognize that qualitative analyses are iterative and recursive by nature.

Prior to initiating data collection, a coding manual containing a beginning list of codes ( Fonteyn, Vettese, Lancaster, & Bauer-Wu, 2008 ; Hsieh & Shannon, 2005 ; Miles et al., 2014 ) derived from the theoretical framework, literature, and the analysis of preliminary data, was developed. Codes are action-oriented words or labels assigned to designated portions (chunks or meaning units) of text reflecting themes or topics that occur with regularity ( Miles et al., 2014 , p. 71). In the coding manual (see example in Table 2 ), themes which were conceptually similar were grouped together using an ethnographic technique of domain analysis ( Spradley, 1980 ). A domain analysis contains a series of themes, a semantic relationship such as “is a component of” or “is a type of,” and the name of the domain. It is read from the bottom up, hence, “Acknowledging the importance of la familia” “is a result of” “cultural expectation.” Between the semantic relationship (is a result of) and the domain name, we inserted a definition of the domain itself (values, beliefs, and activities seen as normative by members of the culture who learn, share, and transmit this knowledge to others).

Example of a Coding Manual.

Note . SES = socioeconomic status.

Reading from the left in Table 2 , codes were given a number and letter for use in marking sections of text. Next, the code name indicating a theme was entered in boldface type with a definition in the code immediately under it. The second column provided an exemplar of each code, along with a notation indicating where it was found in the data, so that coders could recognize instances of that particular code when they saw them.

The coding manual was tested against data gathered in a preliminary study and was revised as codes found to overlap or be missing entirely. We continued to revise it iteratively during the study as data collection and analysis proceeded and then used it to recode previously coded data. Using this procedure, it was used to revisit the data several times.

Each transcribed document was formatted with wide right margins that allowed the investigator to apply codes and generate marginal remarks by hand. Marginal remarks are handwritten comments entered by the investigator. They represent an attempt to stay “alert” about analysis, forming ideas and recording reactions to the meaning of what is seen in the data. Marginal remarks often suggest new interpretations, leads, and connections or distinctions with other parts of the data ( Miles et al., 2014 ). Such remarks are preanalytic and add meaning and clarity to transcripts.

The investigator took sentences or paragraphs in the transcripts and divided them into meaning units, which are segments of text that contain a single idea ( Table 3 ). One or more codes were applied to each meaning unit during first-level coding, which is highly descriptive in nature. In Table 3 , reading from left to right, the first column contains text that has been separated into meaning units by color. The second column lists codes that were applied to each meaning unit, also color coded for clarity. First-level codes are in gerund form: a verb with an “ing” ending that denotes action. Gerunds are used to help the researcher focus on participant behaviors and actions in the transcript. Table 3 is an example of first-level or coarse coding (applying fewer codes to bigger “chunks” of material). Alternatively, individual researchers may choose to code finely (applying more codes to smaller “chunks” of material). Coding is a form of analysis; they “are prompts or triggers for deeper reflection” ( Miles et al., 2014 , p. 73). Because coding is a way to condense data, the researcher may choose to put “chunks” of coded material in large or small groupings, effectively slicing the data in a fine or coarse manner.

Conceptually similar codes were organized into categories (coding groups of coded themes that were increasingly abstract) through revisiting the theory framing the study (asking, “does this system of coding make sense according to the chosen theory?”). Miles et al. (2014) provide many examples for creating, categorizing, and revising codes, including highlighting a technique used by Corbin and Strauss ( Corbin & Strauss, 2015 ) that includes growing a list of codes and then applying a slightly more abstract label to the code, creating new categories of codes with each revision. This is often referred to as second-level or pattern coding, a way of grouping data into a smaller number of sets, themes, or constructs. During the analysis of data, patterns were generated and the researcher spent significant amounts of time with different categorizations, asking questions, checking relationships, and generally resisting the urge to be “locked too quickly into naming a pattern” ( Miles et al., 2014 , p. 69).

During this phase of analysis, pattern codes were revised and redefined in the coding manual and exemplars were used to clarify the understanding of each code. Miles et al. (2014) suggest that software can be helpful during this categorization (counting) step, so lists of observed engagement behaviors were also recorded in Dedoose software ( Dedoose, 2015 ) by code so that frequencies could be captured and analyzed. Despite the assistance of Dedoose, the researcher found that hand sorting codes into themes and categories was best done on paper.

Analytic memos are defined by Miles et al. (2014 , p. 95) as a “brief or extended narrative that documents the researcher’s reflections and thinking processes about the data.” Memos (see Figure 1 as an example) aided in data reduction by tying together different pieces of data into conceptual clusters. Memos were personal, methodological, or substantive in nature. These analytic memos were further analyzed by summarizing and creating additional analytic memos for groups of observations that contained similarities, effectively reducing the data collected through observation. Memoing was conducted throughout the analysis, beginning with data collection and continuing to the dissertation findings to chapter write-up.

Data displays (matrices), or visual representations containing concepts or variables were helpful in analyzing the data ( Table 4 ). Data displays help the investigator draw conclusions through an iterative process whereby collected data are represented in data displays, thereby reducing data and conducting further analysis ( Miles et al., 2014 ). Data displays are used extensively to categorize, organize, and analyze data. Such displays provide an opportunity to combine quantitative and qualitative findings, triangulating data collected by standardized measures, forms, observations, and interviews both within case and cross case. Triangulation refers to the use of more than one approach for investigating the research question in order to enhance confidence in the findings ( Creswell & Plano-Clark, 2007 ; Denzin & Lincoln, 1994 ; Denzin, Lincoln, & Giardina, 2006 ; Sandelowski, 2001 ).

Finally, the data are re-presented in a creative but rigorous way that are judged to best fit the findings ( Miles et al., 2014 ; Sandelowski & Leeman, 2012 ; Stake, 2010 ; Wolcott, 2009 ).

Level 1 Coding With Meaning Units.

Example of an analytic memo used in qualitative description analysis.

Data Matrix.

Note . The CLOX is an executive clock drawing task that tests cognition and was used in this study with the caregiver (CG) and the care recipient (CR). The CG Strain and the CG Gain scores were derived by the researcher through a qualitative content analysis ( Evans, Coon, & Belyea, 2006 ).

Strategies for Ensuring Rigor of Findings

Many qualitative researchers do not provide enough information in their reports about the analytic strategies used to ensure verisimilitude or the “ring of truth” for the conclusions. Miles, Huberman, and Saldana (2014) outline 13 tactics for generating meaning from data and another 13 for testing or confirming findings. They also provide five standards for assessing the quality of conclusions. The techniques relied upon most heavily during a qualitative descriptive study ought to be addressed within the research report. It is important to establish “trustworthiness” and “authenticity” in qualitative research that are similar to the terms validity and reliability in quantitative research. The five standards (objectivity, dependability, credibility, transferability, and application) typically used in qualitative descriptive studies to assess quality and legitimacy (trustworthiness and authenticity) of the conclusions are discussed in the next sections ( Lincoln & Guba, 1985 ; Miles et al., 2014 ).

Objectivity

First, objectivity (confirmability) is conceptualized as relative neutrality and reasonable freedom from researcher bias and can be addressed by (a) describing the study’s methods and procedures in explicit detail, (b) sharing the sequence of data collection, analysis, and presentation methods to create an audit trail, (c) being aware of and reporting personal assumptions and potential bias, (d) retaining study data and making it available to collaborators for evaluation.

Dependability

Second, dependability (reliability or auditability) can be fostered by consistency in procedures across participants over time through various methods, including the use of semistructured interview questions and an observation data collection worksheet. Quality control ( Miles et al., 2014 ) can be fostered by:

deriving study procedures from clearly outlined research questions and conceptual theory, so that data analysis could be linked back to theoretical constructs;

clearly describing the investigator’s role and status at the research site;

demonstrating parallelism in findings across sources (i.e., interview vs. observation, etc.);

triangulation through the use of observations, interviews, and standardized measures to more adequately describe various characteristics of the sample population ( Denzin & Lincoln, 1994 );

demonstrating consistency in data collection for all participants (i.e., using the same investigator and preprinted worksheets, asking the same questions in the same order);

developing interview questions and observation techniques based on theory, revised, and tested during preliminary work;

developing a coding manual a priori to guide data analysis, containing a “start list” of codes derived from the theoretical framework and relevant literature ( Fonteyn et al., 2008 ; Hsieh & Shannon, 2005 ; Miles et al., 2014 ); and

developing a monitoring plan (fidelity) to ensure that junior researchers, especially do not go “beyond the data” ( Sandelowski, 2000 ) in interpretation. In keeping with the qualitative tradition, data analysis and collection should occur simultaneously, giving the investigator the opportunity to correct errors or make revisions.

Credibility

Third, credibility or verisimilitude (internal validity) is defined as the truth value of data: Do the findings of the study make sense ( Miles et al., 2014 , p. 312). Credibility in qualitative work promotes descriptive and evaluative understanding, which can be addressed by (a) providing context-rich “thick descriptions,” that is, the work of interpretation based on data ( Sandelowski, 2004 ), (b) checking with other practitioners or researchers that the findings “ring true,” (c) providing a comprehensive account, (d) using triangulation strategies, (e) searching for negative evidence, and (f) linking findings to a theoretical framework.

Transferability

Fourth, transferability (external validity or “fittingness”) speaks to whether the findings of your study have larger import and application to other settings or studies. This includes a discussion of generalizability. Sample to population generalizability is important to quantitative researchers and less helpful to qualitative researchers who seek more of an analytic or case-to-case transfer ( Miles et al., 2014 ). Nonetheless, transferability can be aided by (a) describing the characteristics of the participants fully so that comparisons with other groups may be made, (b) adequately describing potential threats to generalizability through sample and setting sections, (c) using theoretical sampling, (d) presenting findings that are congruent with theory, and (e) suggesting ways that findings from your study could be tested further by other researchers.

Application

Finally, Miles et al. (2014) speak to the utilization, application, or action orientation of the data. “Even if we know that a study’s findings are valid and transferable,” they write, “we still need to know what the study does for its participants and its consumers” ( Miles et al., 2014 , p. 314). To address application, findings of qualitative descriptive studies are typically made accessible to potential consumers of information through the publication of manuscripts, poster presentations, and summary reports written for consumers. In addition, qualitative descriptive study findings may stimulate further research, promote policy discussions, or suggest actual changes to a product or environment.

Implications for Practice

The qualitative description clarified and advocated by Sandelowski (2000 , 2010 ) is an excellent methodological choice for the healthcare environments designer, practitioner, or health sciences researcher because it provides rich descriptive content from the subjects’ perspective. Qualitative description allows the investigator to select from any number of theoretical frameworks, sampling strategies, and data collection techniques. The various content analysis strategies described in this paper serve to introduce the investigator to methods for data analysis that promote staying “close” to the data, thereby avoiding high-inference techniques likely challenging to the novice investigator. Finally, the devotion to thick description (interpretation based on data) and flexibility in the re-presentation of study findings is likely to produce meaningful information to designers and healthcare leaders. The practical, step-by-step nature of this article should serve as a starting guide to researchers interested in this technique as a way to answer their own burning questions.

Acknowledgments

The author would like to recognize the other members of her dissertation committee for their contributions to the study: Gerri Lamb, Karen Dorman Marek, and Robert Greenes.

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research assistance for data analysis and manuscript development was supported by training funds from the National Institutes of Health/National Institute on Nursing Research (NIH/NINR), award T32 1T32NR012718-01 Transdisciplinary Training in Health Disparities Science (C. Keller, P.I.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the NINR. This research was supported through the Hartford Center of Gerontological Nursing Excellence at Arizona State University College of Nursing & Health Innovation.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

• Corbin J, & Strauss A (2015). Basics of qualitative research: Techniques and procedures for developing grounded theory (4th ed.). Thousand Oaks, CA: Sage. [ Google Scholar ]
• Creswell J (2013). Qualitative inquiry and research design: Choosing among five approaches (3rd ed.). Los Angeles, CA: Sage. [ Google Scholar ]
• Creswell J, & Plano-Clark V (2007). Designing and conducting mixed methods research. Thousand Oaks, CA: Sage. [ Google Scholar ]
• Dedoose. (2015). Version 6.1.18, web application for managing, analyzing, and presenting qualitative and mixed method research data Los Angeles, CA: SocioCultural Research Consultants, LLC; Retrieved from www.dedoose.com [ Google Scholar ]
• Denzin N, & Lincoln Y (1994). The handbook of qualitative research. New York, NY: Sage. [ Google Scholar ]
• Denzin N, Lincoln Y, & Giardina M (2006). Disciplining qualitative research. International Journal of Qualitative Studies in Education, 19, 769–782. [ Google Scholar ]
• Evans BC, Belyea MJ, Coon DW, & Ume E (2012). Activities of daily living in Mexican American caregivers: The key to continuing informal care. Journal of Family Nursing, 18, 439–466. doi:410.1177/1074840712450210. Epub 1074840 712452012 Jun 1074840712450226 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Evans BC, Belyea MJ, & Ume E (2011). Mexican-American males providing personal care for their mothers. Hispanic Journal of Behavioral Sciences, 33, 234–260. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Evans B, Coon D, & Belyea M (2006). Worry among Mexican American caregivers of community-dwelling elders. Public Health Nursing, 23(3), 284–291. doi: 10.1111/j.1525-1446.2006.230312.x [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Evans BC, Coon DW, & Ume E (2011). Use of theoretical frameworks as a pragmatic guide for mixed methods studies: A methodological necessity? Journal of Mixed Methods Research, 5, 276–292. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Fonteyn ME, Vettese M, Lancaster DR, & Bauer-Wu S (2008). Developing a codebook to guide content analysis of expressive writing transcripts. Applied Nursing Research, 21, 165–168. doi: 10.1016/j.apnr.2006.08.005 [ DOI ] [ PubMed ] [ Google Scholar ]
• Hsieh H, & Shannon S (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15, 1277–1288. [ DOI ] [ PubMed ] [ Google Scholar ]
• Jiggins Colorafi K (2015). Patient centered health information technology: Engagement with the plan of care among older adults with multi-morbidities. Retrieved from ProQuest. [ Google Scholar ]
• Lincoln Y, & Guba E (1985). Naturalistic inquiry. New York, NY: Sage. [ Google Scholar ]
• Miles M, Huberman M, & Saldana J (2014). Qualitative data analysis: A methods sourcebook (3rd ed.). Thousand Oaks, CA: Sage. [ Google Scholar ]
• Munhall P, & Chenail R (2008). Qualitative research proposals and reports: A guide (3rd ed.). Boston, MA: Jones and Bartlett. [ Google Scholar ]
• Neergaard MA, Olesen F, Andersen RS, & Sondergaard J (2009). Qualitative description - the poor cousin of health research? BMC Medical Research Methodology, 9, 52. doi: 10.1186/1471-2288-9-52 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sandelowski M (2000). Whatever happened to qualitative description? Research in Nursing & Health, 23, 334–340. [ DOI ] [ PubMed ] [ Google Scholar ]
• Sandelowski M (2001). Focus on research methods. Real qualitative researchers do not count: The use of numbers in qualitative research. Research in Nursing & Health, 24, 230–240. [ DOI ] [ PubMed ] [ Google Scholar ]
• Sandelowski M (2004). Counting cats in Zanzibar. Research in Nursing & Health, 27, 215–216. [ DOI ] [ PubMed ] [ Google Scholar ]
• Sandelowski M (2010). What’s in a name? Qualitative description revisited. Research in Nursing & Health, 33, 77–84. doi: 10.1002/nur.20362 [ DOI ] [ PubMed ] [ Google Scholar ]
• Sandelowski M, & Leeman J (2012). Writing usable qualitative health research findings. Qualitative Health Research, 22, 1404–1413. [ DOI ] [ PubMed ] [ Google Scholar ]
• Spradley J (1980). Patricipant observation. Belmont, CA: Wadsworth Cengage Learning. [ Google Scholar ]
• Stake R (2010). Qualitative research: Studying how things work. New York, NY: The Guilford Press. [ Google Scholar ]
• Wolcott H (2009). Writing up qualitative research (3rd ed.). Los Angeles, CA: Sage. [ Google Scholar ]
• View on publisher site
• PDF (406.2 KB)
• Collections

Similar articles

Cited by other articles, links to ncbi databases.

• Download .nbib .nbib
• Format: AMA APA MLA NLM

Add to Collections