Research Applications of Nursing Informatics

Problems whose sole purpose is merely to compare two sets of data

Problems that result in a YES or NO answer

Those that deal with ethical or moral questions

Questions that do not require a mental struggle

When a machine can take place of a researcher to provide answers to questions

Metaphysical questions where it is impossible to collect publicly verifiable data and information in order to answer the question.

What are NOT research problems?

evidence-base practice

“The process of shared decision making between practitioner, patient, and others significant to them based on research evidence, the patient’s experiences and preferences, clinical expertise or know-how, and other available robust sources of information.”

•research the problem and use evidence-based

resources;

•critically select electronic and non-electronic

references;

•consolidate the research findings, combine and

compare the conclusions; and

•present the findings, and propose a solution

nurses should:

professional online databases

•A source of online information that is generally

invisible to all Internet users except those with

professional or academic affiliations, such as

faculty, staff, and students

databases

specific or general — act as collection

points such as abstracts and articles from many

different journals

CINAHL (Cumulative Index to Nursing and Allied Health Literature)

includes information from all aspects of allied

health, nursing, alternative medicine, and community

medicine.

Cumulative Index to Nursing and Allied Health Literature

meaning of CINAHL

MEDLINE

contains references to more than 22 million

journal articles and is maintained and produced by the

National Library of Medicine

search engines

allow users to surf the Web

and find information on nearly anything.

search engines

•Vast amounts of unsubstantiated information

are likely to be uncovered

web address

indicates the entity responsible for creating

the website — not a sole deciding factor when choosing

to trust information

electronic library catalog

• Common in HEIs

• Assistance from library professionals — some are

unused to working completely online and may be

intimidated by an e-catalog

positivism

believes that there is an objective reality

numbers

represent objectivity, neutrality

Paper-and-Pencil Questionnaire (PPQ)

manual encoding into the compute

automated devices

Tools like “Smart Caps”

online survey tools

SurveyMonkey, E-Surveys Pro,

and Qualtrics

OCR & Scanning Software:

SurveyMonkey, E-Surveys Pro,

and Qualtrics

direct device measurement & digitized imaging data

physiological data capture

direct device measurement

Physiological data (e.g., heart

rhythm, rate, fluid/electrolytes) is

directly captured from patient

devices and integrated into

(electronic) hospital records.

digitized imaging data

Measurements from

neurological, cardiovascular,

and cellular imaging are now

digitized and entered directly

into computer programs for

analysis.

big data

resulted from the improved capacity to capture

and collect data

data mining

is the process of discovering patterns,

trends, anomalies, and correlations within large

datasets.

data coding

Variables (numerical or not) are

collected for numerical analysis —

ergo, there is a need to transform

(i.e., quantify) data.

descriptive & exploratory analyses

hypothesis testing

estimation of confidence intervals

model building (multivariate analysis)

path analysis & structural equation modeling

combined approaches

Key Goals of Data Analysis in Nursing Research

descriptive & exploratory analyses

Summarizing

and understanding the basic characteristics of

data.

hypothesis testing

Statistically testing

relationships or differences proposed by

hypotheses.

estimation of confidence intervals

Quantifying

the uncertainty around an estimate.

model building (multivariate analysis)

Developing statistical models to understand

complex relationships among multiple variables.

path analysis & structural equation modeling

Advanced techniques for examining direct and

indirect effects among variables in complex

theoretical models.

combined approaches

Many studies (e.g.,

intervention studies, quality improvement, patient

outcomes) utilize multiple types of analysis.

IBM SPSS Statistics 24

Widely used, user-friendly,

with layout and export efficiencies.

statistical analysis services (SAS)

Robust and

powerful for complex analyses.

STATA 15

Another comprehensive statistical

software.

“R” software

▪Free, open-source language and computing

environment

▪Performs basic to advanced statistical and graphical

techniques.

▪Extensible, with growing use in cutting-edge research.

qualitative data analysis

often occurs concurrently

with data collection, reflexively and iteratively.

ongoing & iterative analysis

• There is no clear end to the collection process; it's a

continuous process.

• The focus is on providing a general understanding of

the phenomenon from the experiences of different

individuals

computer programs

mechanical tools that aid in

manipulating both quantitative

and qualitative data.

interviews

often audiotaped,

then transcribed into word

processing programs.

audacity

An open-source, free audio-recording

package for editing and exporting voice data for

analysis.

survey monkey

Simple online survey tools can now

export participants' free text data into qualitative

software packages.

historical challenges

• Relied on narrative notes,

often from audio

transcriptions.

• Manual coding was time-

consuming, involving

thousands of pages and

physical manipulation

(scissors, tape).

Specialized Qualitative Data Analysis (QDA) Software

•Helps researchers establish code indexes and identify

relationships among categories.

• Facilitates coding and recoding large amounts of

data, encouraging experimentation with data

organization.

role of computerization in analysis

• Extremely helpful for handling large amounts of

qualitative data.

• Crucially, computer applications are management

tools, not analytical ones.

• Synthesis and interpretation of data remain the

researcher's work.