NURSING RESEARCH

Research
- systematic, disciplined investigation
- use of scientific method

Nursing research
Purpose: to develop evidence on issues of the nursing profession
- E.g. Nursing practice, education, administration.
Goal: improve QOL (quality of life) of patients

Clinical nursing research

Purposes: to guide bedside practice

BASIC CONCEPTS

Sources of Evidence
1) Tradition 4) Trial and error
- beliefs are truths - risky, may cause injury
- based on customs
- not always correct 5) Intuition: hunch, instinct

2) Authority 6) Logical Reasoning

- expert in field a) Inductive reasoning
- not always correct - specific to general
b) Deductive reasoning
3) Clinical Experience - general to specific
- based on nurse’s observations
- may be biased 7) Disciplined Research
- rigorous, systematic, evidence-based

Basic vs. Applied Research

1) Basic: increase knowledge
2) Applied: find solutions to problems, higher level than basic

Variable
1) Independent variable (IV): presumed cause
2) Dependent variable (DV): presumed effect, aka outcome variable
IV → DV
Presumed cause Presumed effect

3) Confounding/Extraneous Variables
- contaminating factors

Relationships
- connection between two or more phenomena

a) Causal relationship
- cause-and-effect (100% sure)

b) Associative relationship
- Change in X tends to change Y (just a probability)
1
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
 Quantitative Research Qualitative Research

Empirical: observed, not hunch Understanding human experience

Systematic: logical, planned Flexible, evolving procedures

Quantitative data: numbers, statistics Qualitative data: narrative, subjective

validity, reliability biases Naturalistic setting

Deductive reasoning Inductive reasoning

5 Research Phases (Quantitative Study) (CDEAD)

1) Conceptual Phase
a) Define the problem.
b) Review the related literature.
c) Undertake clinical fieldwork.
– observing current practices in hospital
- gathering background data
d) Define the framework.
e) Formulate hypotheses.

2) Design and Planning Phase

a) Select a research design.
b) Develop intervention protocols.
– if treatment is involved
- only for experimental design
c) Identify the population.
d) Design the sampling plan.
e) Specify data collection methods.
f) Develop methods to safeguard subjects. – ethical considerations
g) Finalize the research plan. - proposal defense

3) Empirical Phase
a) Collect the data.
b) Prepare data for analysis. - encoding

4) Analytic Phase
a) Analyze the data. (numbers)
– statistician
b) Interpret the results. (words)

5) Dissemination Phase
a) Communicate the findings.
b) Utilize the findings in practice.

2
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
Ethics in Research
Codes of Ethics - protection of research subjects
Priority: Safety

Codes of Ethics
1) Nuremberg Code 3) Belmont Report
- Informed consent is absolute - basic ethical principles
requirement
- ONLY ADULTS can give consent 4) Code of Ethics for Nurses
2) Declaration of Helsinki - BON Board Resolution No. 220, Series
- ✓ children and vulnerable populations of 2004
- less strict - preserve health at all cost

Ethical Principles (from Belmont Report)

1) Beneficence: Do good (e.g. medications) Right: Self-determination

Non-maleficence: Do no harm (e.g. asepsis)
4) Veracity: complete information about the
Right: Freedom from Harm study
- Right: Full disclosure
2) Justice: equal risks and benefits
Right: Fair Treatment 5) Confidentiality: data not revealed
- Anonymity: identity cannot be linked to
3) Autonomy data
- voluntary decision - Right: Privacy

Informed Consent
- ✓accept or X decline participation voluntarily
- protects right to self-determination

Informed Consent (VICS)

- Voluntary
- Informed: fully understood
- Competence: 18 y/o and above (legal age), coherent
- Signature: (1) Participant, (2) witnesses

Ethical Guidelines
< 7 y/o – No assent needed, as long as no dissent (Assent: agreement of minor)
7 to <12 y/o – Verbal assent
12 to <15 y/o – Assent form
15 to <18 y/o – Informed consent form w/ co-sign of parents
18 y/o and up – Informed consent form

Institutional Review Board

- ensures that studies conduct ethical procedures
- Actions: ✓approve,  require modifications, X disapprove

3
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
CHAPTER 1: INTRODUCTION

Research Problem
- “What is wrong?”
- Curiosity and interests

Statement of Purpose
- “What do you want to happen?”
- aka. Goals (general) and Objectives (specific)

Research Question
- question to be answered in addressing the research problem
- Basic form (PIRD):
In (1. Population), is (2. Independent variable [IV]), (3. Relationship) with (4. Dependent
variable [DV])?”
- E.g.] Among older adults (population), is high sugar intake (IV) associated with
(relationship) diabetes mellitus (DV)?

Evidence-based Practice (EBP)

- Application of findings to clinical practice.
- Starts with clinical question
- Cochrane: best source of EBP
- Clinical Question (PICO)
Population, Intervention, Comparison, Outcome
E.g.: Among cachexic patients, is megestrol more effective than glucocorticoids in causing
weight increase?
Population is cachexic patients; the intervention and comparison is appetite stimulants
(megestrol and glucocorticoids); and the outcome is weight increase.

Hypothesis
- predicted relationship between two or more variables
- Formulated at Conceptual phase
- Hypothesis testing: thru statistical analysis

Directional vs Nondirectional
Directional
- shows direction of the relationship
- “more”, “less”
- X, Y: direct
- X, Y: inverse
- E.g. The telemedicine for pediatric cancer patients is effective in increasing their health status.

Nondirectional hypothesis
- No direction of the relationship
- “related to, associated with”
- E.g. The telemedicine for pediatric cancer patients is associated with their health status.

4
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
Research/Alternative vs Null hypothesis
- Research/Alternative hypotheses (Ha) - Null hypotheses (Ho)
- Actual relationship. - No relationship.
- There is a significant difference. -There is no significant difference.

CHAPTER 2: REVIEW OF RELATED LITERATURE

Purposes
-Identify gaps in knowledge
-Identify similarities and differences with other studies
-Highlight significance of study

Characteristics (CUP B)
-Comprehensive
-Up-to-date: at most 5 years ago
-Paraphrased: X copy, paste
-Balanced: no bias

Types of Sources
1) Primary source: Reports of original researchers
2) Secondary source: Reports by someone other than original researcher

Primary > Secondary

Search Strategies
1) Bibliographic databases
- Use of keywords to search

2) Ancestry approach
- “footnote chasing”
- older studies (ancestors)  references  newer study

3) Descendancy approach
- older study → citation → recent studies (descendants)

Theoretical vs. Conceptual Framework

Theory: abstract explanation
Conceptual model: visual representations of relationship of phenomena

Operational Definition of Terms

- how terms mean in the study
- for uniformity of data
e.g., Weight: measured using spring scale in kg; fully undressed, every 8 am.

5
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
CHAPTER 3: METHODOLOGY

Basic Concepts
1) Bias
- influence that produces an error
- decreases validity
- minimize; cannot be avoided
- e.g. sampling bias

2) Validity
- You measure what you intend to measure.

A) Internal Validity
- Does the independent variable truly influence the dependent variable?

B) External Validity
- Different settings, Same results
- Quantitative study: Generalizability
- Qualitative study: Transferability

3) Reliability
- Same setting, Same results
- consistency

* All valid are reliable. NOT all reliable are valid.

*  reliability and validity,  scientific merit (Kapani-paniwala)

3) Blinding
- disguising information
- to prevent biases due to knowledge of treatment
- Single-blind: researchers OR participants don’t know
Double-blind: both researchers AND participants don’t know
Double-blind > Single-blind

4) Reflexivity
- reflecting critically on the self
- used in qualitative studies
6
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
Types of Research Designs

Quantitative Designs
I.) True Experimental (or Randomized Controlled Trial or Clinical Trial)

- 3 Features (CRM)

a.) Control
- Control group (for comparison)
- no intervention/ alternative intervention/ placebo
- Placebo: false intervention, no value

b.) Randomization
- Random assignment to experimental and control groups
- EG and CG are EQUIVALENT. Cancels confounding variables.  Bias

c.) Manipulation
- intervention or treatment of experimental group

II.) Quasi-experimental
-Absence of randomization
-Experimental and Control groups are NON-EQUIVALENT.
-Experimental > Quasi-experimental

III) Non-experimental
-Researchers: by-standers
- When independent variables cannot be manipulated

1) Descriptive
- observe, describe, and document variables

2) Correlational
- association between two variables
- Correlation  causation
- Descriptive correlational: describes relationships among variables.

3) Prospective/ Cohort
- follow to the future
- Sample → IV (Exposed/Not exposed) → DV (affected/not affected)

Time dimension: Longitudinal

- data collection: multiple points
- Explains trends over time

4) Retrospective/ Case-control
- look back in the past
- IV (Exposed/Not Exposed)  DV (Case/Control)  Sample

Time dimension: Cross-sectional

- data collection: one point in time
- Cannot explain trends over time
7
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
 *Prospective > Retrospective
Why? Prospective design can be controlled/monitored!

Qualitative Designs
-identify themes and categories
-flexible and elastic; evolves throughout study
-Triangulation: use of multiple sources; strengthens evidence.

1) Ethnography
-Cultural patterns, lifeways, and experiences
-2 Perspectives
o Etic: outsider’s view (researcher)
o Emic: insiders’ view (participant)

2) Phenomenology
-lived experience
-discovering meaning

3) Grounded theory
-understanding social processes
-e.g. the process of practicing spirituality for health promotion of Filipinos

4) Case Studies
-in-depth study of a single entity
- E.g. Case presentation

5) Participatory Action Research (PAR)

-Collaboration: researchers x participants
-Tasks: (1) conduct research, (2) empower people
-Type: Action research; Design: Qualitative

Threats to Internal Validity

a) Selection threat: When subjects not randomized.

b) History threat/Co-intervention bias
- Independent variable + previous/ concurrent intervention → Dependent variable
c) Maturation threat: Due to passing of time
d) Mortality: A.k.a. Attrition, Drop-out or deaths

Sampling
-Selecting a portion of the population (sample)
-Population (N), sample (n)
-Goal: Representativeness
-Whole population: Census

Population
1) Target population: entire population of interest
2) Accessible population: part of target population accessible

8
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
 *Target population > Accessible population > Sample*

Strata
- Specific characteristic
-E.g., nursing students. Strata: 1st year, 2nd year, 3rd year, 4th year

Sample size
1) Quantitative
Slovin’s formula
𝑵
𝒏=
𝟏 + 𝑵𝒆𝟐
e=margin of error

2) Qualitative
-Data saturation: data repetitive

Types of Sampling

Non-probability Sampling
- Non-random
- Not all have the chance to be selected
- Low representativeness

1) Convenience 3) Quota
- most conveniently available people - Strata; non-random selection
- Biased, weakest, but most common
4) Purposive/judgmental
2) Snowball - Hand-picking samples
- Referrals - Inclusion and exclusion criteria

Probability Sampling
- Random
- All have chance to be selected.
- High representativeness
- Probability sampling > Non-probability sampling

1) Simple Random Sampling 3) Stratified Random Sampling

- Sampling frame: list of population - Strata; random
- E.g. draw lots
4) Cluster/ Multistage
2) Systematic Random Sampling - Successive random sampling
- Sampling interval (k); k=N/n - Geographic locations

9
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
Data Collection Methods
1) Observation
- Rating scales, checklists
- Participant observation: observe people in natural environment
- Problem: Hawthorne effect
- Knowledge of being observed → Change in behavior → Inaccurate results

2) Self-reports
- Responses to questions, most common
- Quantitative: questionnaire
- Qualitative
a) Semi-Structured interview – list of topics/questions
b) Unstructured interview – no previous knowledge
c) Focus group discussions (FGD) – 5 to 10 people

3) Biophysiologic measures
- Assessment of clinical variables
a) In vivo: within humans
e.g., Blood pressure, Temp, PR, RR, O2 saturation
b) In vitro: extracted from humans
e.g., ABGs, CBC, urine

Levels of Measurement (NOIR)

1) Nominal
- Lowest level 3) Interval
- Words only, no numbers - Zero is not absolute
- E.g. sex, blood type - E.g. Temperature (°C) = 0 °C is not
absence of heat
2) Ordinal
- Ranking 4) Ratio
- Numbers mean order - Absolute zero
- E.g. Top 1, Top 2, Top 3 - Highest level

CHAPTER 4: DATA ANALYSIS AND INTERPRETATION

Descriptive Statistics
-To synthesize and describe data

Measures of Central Tendency

- Mean: average
- Median: number that divides distribution in half
- Mode: most frequent number

Inferential Statistics
- To make interpretations about the population
- Used for Hypothesis Testing

10
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
Type I vs Type II errors
Type I error
-Akala mo meron (significant difference), pero wala, wala, wala!
Type II error
-Akala mo wala, pero meron, meron, meron!

1) Correlation
Pearson’s r value
Interpretation:
0 – no relationship
0.1 to 0.4 – low
0.41 to 0.5 – moderate
0.51 to 1.0 – high

(+) positive correlation: direct

(-) negative correlation: inverse

2) Chi-squared (X2) test

- Differences of the proportion (%) of frequencies

3) t-test
- Differences of means of 2 groups.

Independent vs. Dependent t-test

a) Independent: Group 1 and Group 2 are different groups (magkahiwalay)
b) Dependent/Paired: Same group (magkasama)

4) Analysis of Variance (ANOVA)

- Differences of means of 3 or more groups.

Systematic Reviews
-Integration of multiple studies
-Benefits
o Power – detects true relationships
o Objectivity – minimized bias
o Precision – accurate conclusions

Types
1) Meta-analysis: quantitative studies
2) Meta-synthesis: qualitative studies

11
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)
Abstract
-Brief study description (~150-300 words)
-Beginning of the article.

Abstract Format (IMRaD)

Introduction
Method
Results
and
Discussion

RESEARCH DISSEMINATION
1) Presentations
- Oral Presentation
- Key features of the study
- Big group
- 10 to 20 minutes
- Poster presentation
- Visual display + discussion
- Small group

2) Publication (Journal Articles)

- Peer-review: Blind review of manuscripts by 2 or more researchers
Blind: Names not revealed.
- Evaluates strengths and weaknesses
- Decision: / Accept, o Suggest revisions, X reject

Levels of Research Evidence

Level I: Systematic review of Randomized Controlled Trials (highest)
Level II: Single RCT
Level III: Systematic review of correlational studies
Level IV: Single correlational study
Level V: Systematic review of descriptive studies
Level VI: Single descriptive studies
Level VII: Expert opinion (lowest)

12
Prepared by: Raymund Kernell B. Mañago, RN
Source: Polit and Beck’s Essentials of Nursing Research (2010)