Midwife Licensure Exam Research & Evidence-Based Practice — Nursing Research Process & Evidence-Based PracticeSummary

Nursing research is a systematic, controlled inquiry designed to develop, refine, and expand nursing knowledge through rigorous investigation of phenomena relevant to the discipline. Its overarching purposes include identifying healthcare problems, describing patient experiences and phenomena, explaining cause-and-effect relationships, predicting outcomes, and controlling or managing nursing interventions to improve quality of care. In the Philippine healthcare context, nursing research supports clinical decision-making, informs policy at facility and national levels, strengthens the profession's autonomy and credibility, and ensures that practice is grounded in evidence rather than tradition alone. The Board of Nursing mandates that all registered nurses understand research fundamentals to uphold RA 9173's standards.

Quantitative research follows an orderly, logical sequence: (1) Identify and state the problem—defining a clear gap in knowledge and its clinical or professional significance; (2) Review the literature systematically—synthesizing what is already known and pinpointing the exact gap the study will address; (3) Formulate the theoretical or conceptual framework—selecting the lens (theory, model, or conceptual scheme) through which the researcher views the phenomenon; (4) State research questions, objectives, or hypotheses—specifying what will be investigated and what relationships or differences are expected; (5) Select the research design and methodology—choosing the approach that best answers the question; (6) Identify the population and sample and select a sampling method—defining who will be studied and how they will be chosen; (7) Collect data—gathering information using valid and reliable instruments; (8) Analyze data—applying appropriate statistical methods; (9) Interpret findings—making sense of results and drawing conclusions; (10) Communicate and disseminate—publishing or presenting findings and recommending applications. This systematic approach ensures rigor, reproducibility, and the integrity of the knowledge produced.

In quantitative research, variables are attributes or characteristics that vary and can be measured. The independent variable (IV) is the presumed cause, treatment, or intervention that the researcher manipulates or introduces. The dependent variable (DV) is the outcome or effect—what is measured to see if the intervention worked. For example, in a study examining whether pain education (IV) reduces postoperative pain scores (DV), pain education is manipulated and pain scores are measured. A hypothesis is a testable prediction about the relationship between variables. The null hypothesis (H₀) states that there is no relationship, no difference, or no effect—it is the conservative position that statistical testing attempts to reject. The alternative or research hypothesis (H₁) states the expected relationship or difference. In NLE questions, you may see: H₀: There is no difference in patient satisfaction between nurses who use Protocol A versus Protocol B. H₁: There is a difference in patient satisfaction between the two protocols.

Quantitative research measures variables numerically, tests hypotheses, and seeks objective, generalizable results that can be applied broadly. It employs deductive logic—starting from theory or hypothesis and testing it with data. Qualitative research explores meaning, experience, and process through narrative, descriptive data (words, not numbers). It uses inductive logic—starting with observation and building theory from patterns in the data. Qualitative traditions include: (1) Phenomenology—examining the lived experience of individuals (e.g., the experience of grief after loss); (2) Grounded theory—developing new theory from data systematically collected and analyzed (e.g., how families adapt to a chronic illness); (3) Ethnography—studying the culture, beliefs, and practices of a group (e.g., healing practices in a specific Filipino community); (4) Case study—in-depth examination of a single case or small number of cases. Neither approach is superior; each answers different questions. Quantitative answers 'how much' and 'is there a relationship?'; qualitative answers 'what is the experience?' and 'why do people behave this way?'

Research designs are classified by their ability to infer cause and effect, from strongest to weakest: EXPERIMENTAL DESIGNS have three hallmarks: (1) Manipulation of the independent variable—the researcher actively introduces the intervention; (2) A control group—a comparison group that does not receive the intervention; (3) Randomization—random assignment of subjects to intervention or control groups, reducing bias. The Randomized Controlled Trial (RCT) is the gold standard for establishing cause and effect. QUASI-EXPERIMENTAL DESIGNS have manipulation but lack either randomization or a control group (or both), making causal inference weaker but studies more feasible in real clinical settings. NON-EXPERIMENTAL/OBSERVATIONAL DESIGNS do not manipulate variables and observe or measure naturally occurring phenomena: (1) Descriptive—describes characteristics, frequencies, or patterns (e.g., 'What are the anxiety levels in pre-operative patients?'); (2) Correlational—examines relationships between variables without manipulation (e.g., 'Is there a relationship between sleep deprivation and medication errors?'); (3) Cohort/Prospective—follows groups forward in time to see who develops an outcome (strong for establishing causation in observational studies); (4) Case-control/Retrospective—compares those with an outcome to those without, looking backward (useful when the outcome is rare); (5) Cross-sectional—data collected at one point in time (useful for prevalence studies). Remember: correlation does not prove causation.

The population is the entire group of individuals to whom the research question applies. The target population is the group to which the researcher wishes to generalize findings (e.g., all Filipino patients with Type 2 diabetes). The accessible population is the portion actually available for study (e.g., patients with Type 2 diabetes in a particular provincial hospital). A sample is a subset of the population that is actually studied. Sampling is the process of selecting the sample. Effective sampling reduces cost and time while maintaining the ability to draw valid conclusions. PROBABILITY (RANDOM) SAMPLING methods give every member of the population a known, independent chance of selection and support generalizability: (1) Simple random—using a random number generator or lottery to select participants; (2) Systematic—selecting every kth person from a list (e.g., every 5th patient on the census); (3) Stratified random—dividing the population into strata (subgroups) based on a characteristic, then randomly sampling within each stratum to ensure representation (e.g., stratifying by age, sex, or disease severity); (4) Cluster/Multistage—randomly selecting geographic clusters or groups, then sampling within them (practical for geographically dispersed populations). NON-PROBABILITY SAMPLING does not give all members an equal chance and is weaker for generalizability but is often practical in qualitative work: (1) Convenience—selecting readily available subjects (weakest, prone to bias); (2) Quota—convenience sampling with preset numbers in each subgroup; (3) Purposive/Judgmental—deliberately selecting subjects who possess a particular characteristic or experience; (4) Snowball/Network—existing subjects refer new participants, useful for hard-to-reach populations (e.g., undocumented immigrants). Larger, well-selected samples reduce sampling error and increase statistical power.

Data collection methods include questionnaires, structured or unstructured interviews, observation, biophysiologic measurements (e.g., blood pressure monitors, pulse oximeters), and review of existing records or databases. Instrument quality is determined by two essential properties: VALIDITY is the extent to which an instrument measures what it is intended to measure. Types include: (1) Content validity—expert judgment that the instrument comprehensively covers the domain (e.g., does a pain scale capture sensory, emotional, and behavioral dimensions?); (2) Construct validity—evidence that the instrument measures the theoretical construct it claims to measure; (3) Criterion validity—the instrument correlates with an external standard or gold standard (e.g., does a new screening tool for depression correlate with clinician diagnosis?). RELIABILITY is the consistency and reproducibility of measurements. Types include: (1) Test-retest reliability—the same instrument yields consistent results when administered to the same subjects at different times (e.g., administering a coping scale twice within a week); (2) Internal consistency—items within an instrument correlate with each other, measured by Cronbach's alpha (range 0–1, with >0.70 generally acceptable); (3) Interrater reliability—different raters or observers score the same phenomenon similarly (important for observational tools). A critical principle: an instrument can be reliable (consistent) without being valid (correct), but it cannot be valid without being reliable. Always assess both before using an instrument in practice or accepting study findings.

Research ethics are built on foundational principles from the Belmont Report: Respect for Persons (autonomy and self-determination), Beneficence (maximizing benefits), and Justice (fair distribution of benefits and burdens). In nursing, we add Nonmaleficence (do no harm) and Fidelity (honoring trust). KEY ETHICAL PROTECTIONS include: (1) Informed Consent—participants must voluntarily agree to participate with full understanding of the purpose, procedures, risks, benefits, and right to withdraw without penalty at any time. Consent is a process, not just a signature. (2) Confidentiality and Anonymity—researchers protect participant identity and data. In the Philippines, the Data Privacy Act of 2012 (RA 10173) mandates protection of personal information. (3) Protection from Harm—researchers must minimize risk and ensure benefits outweigh risks. Physical, psychological, social, and economic harms must be considered. (4) Right to Self-Determination and Full Disclosure—participants decide about their involvement based on complete, honest information. (5) VULNERABLE POPULATIONS require additional safeguards: Children—parental consent plus the child's assent (agreement appropriate to their developmental level); Pregnant women and fetuses—extra protections due to potential harm; Prisoners—risk of coercion; Cognitively impaired individuals—may not understand consent; Critically ill patients—power imbalance with healthcare providers. ETHICS REVIEW: All research involving human subjects must be reviewed and approved by an Institutional Review Board (IRB) or Ethics Review Committee before it begins. In the Philippines, oversight follows the National Ethical Guidelines for Health and Health-Related Research (established by PHREB and DOST). As a nurse, you protect participants' rights by reporting any unethical conduct and ensuring protocols are followed during data collection.

Data can be classified by measurement level, which determines what statistics are appropriate: NOMINAL—categories with no inherent order (sex: male/female; blood type: A/B/AB/O; marital status: single/married/divorced). Only the mode and frequency distribution are appropriate. ORDINAL—ranked categories without equal intervals (pain scale 0–10; Likert scale: strongly disagree to strongly agree; disease severity: mild/moderate/severe). Median and mode are appropriate; mean is not because intervals are unequal. INTERVAL—ordered categories with equal intervals between values but no true zero (temperature in °C; calendar year). Mean, median, mode are all appropriate, and interval-level statistics can be used. RATIO—equal intervals with a true zero point (weight in kg, pulse in beats per minute, blood pressure in mmHg, patient age, income in pesos). All statistics are appropriate, and ratios between values are meaningful (a 100 kg patient weighs twice as much as a 50 kg patient). DESCRIPTIVE STATISTICS summarize sample data: MEASURES OF CENTRAL TENDENCY describe the typical value: Mean (average)—sum of values divided by number of values; best for interval/ratio data and normally distributed data but sensitive to outliers. Median (middle value)—best for ordinal data or skewed data; unaffected by outliers. Mode (most frequent value)—only meaningful measure for nominal data. In a perfectly normal (bell curve) distribution, mean = median = mode. MEASURES OF VARIABILITY describe spread: Range—difference between highest and lowest value; simple but affected by outliers. Standard Deviation (SD)—how far values typically deviate from the mean; used for interval/ratio data. Variance—SD squared. In a normal distribution: ~68% of values fall within ±1 SD, ~95% within ±2 SD, and ~99.7% within ±3 SD. This is called the empirical rule or 68-95-99.7 rule.

Inferential statistics allow researchers to generalize from a sample to a population and test hypotheses about relationships or differences. THE P-VALUE is the probability that an observed result occurred by chance alone if the null hypothesis is true. The conventional cutoff is p < 0.05, meaning if the null hypothesis were true, there is less than a 5% probability of observing a result this extreme by chance. Results with p < 0.05 are declared statistically significant—we reject the null hypothesis and conclude there is likely a real relationship or difference. Important: statistical significance does NOT automatically mean clinical significance; a small effect might be statistically significant in a large sample but clinically trivial. COMMON HYPOTHESIS TESTS: t-test—compares means of two groups (independent t-test for unrelated groups; paired t-test for related groups, such as before-and-after measurements). ANOVA (Analysis of Variance)—compares means of three or more groups. Chi-square (χ²)—tests associations between categorical variables (e.g., Is there an association between gender and job satisfaction?). CORRELATION COEFFICIENT (r)—measures the strength and direction of a linear relationship between two variables. Range: −1 to +1. Sign indicates direction: positive r = as one variable increases, the other increases; negative r = as one variable increases, the other decreases. Magnitude indicates strength: r near 0 = weak/no linear relationship; r near ±0.5 = moderate; r near ±1 = strong. For example, r = −0.85 between sleep deprivation and cognitive function indicates a strong negative relationship. ERRORS IN HYPOTHESIS TESTING: Type I error (alpha)—rejecting a true null hypothesis; a false positive (e.g., concluding a treatment works when it doesn't). Type II error (beta)—failing to reject a false null hypothesis; a false negative (e.g., concluding a treatment doesn't work when it does). Researchers try to minimize both, though Type I error is typically controlled more strictly.

Evidence-Based Practice is a clinical problem-solving approach that integrates three components: (1) the best available research evidence, (2) the nurse's clinical expertise and experience, and (3) the patient's values, preferences, and circumstances. EBP moves beyond tradition-based practice ('we've always done it this way'), opinion-based practice ('the expert says so'), and intuition alone. Instead, it systematically seeks and appraises the strongest available evidence and applies it thoughtfully in the clinical context. In the Philippine healthcare setting, where resources are often limited and best practices must be adapted to local contexts, EBP is essential for optimizing outcomes and efficiency. RA 9173 expects registered nurses to engage in evidence-based practice as part of their professional competence. EBP is NOT the same as research—research generates new knowledge; EBP applies existing best evidence to answer specific clinical questions. EBP is also distinct from Quality Improvement (QI)—QI uses data to improve a particular local process (e.g., reducing catheter-associated infections in your ICU), not to generate generalizable knowledge.

EBP is executed through five sequential steps: STEP 1: ASK—Frame a focused clinical question using the PICO(T) framework: P = Population (Who? What is the specific group?); I = Intervention (What is the treatment or action being considered?); C = Comparison (What is the alternative or standard care?); O = Outcome (What do we want to achieve?); T = Time (Over what timeframe?). Example: 'In post-operative cardiac patients (P), does early ambulation (I) compared with bed rest (C) reduce the incidence of deep vein thrombosis (O) within 7 days post-op (T)?' A well-formed question guides your search. STEP 2: ACQUIRE—Systematically search for the best available evidence using databases (PubMed, CINAHL, Google Scholar, Philippine nursing journals). Look for primary sources (original research) and secondary sources (systematic reviews, meta-analyses, clinical guidelines). Cast a wide net initially, then narrow using inclusion/exclusion criteria. STEP 3: APPRAISE—Critically evaluate the evidence for validity, reliability, applicability, and impact. Ask: Is the study design appropriate? Are instruments valid and reliable? Are the findings statistically and clinically significant? Can results apply to my patient population? Would the intervention be feasible and acceptable in my setting? Consult the evidence hierarchy—systematic reviews and RCTs are strongest. STEP 4: APPLY—Integrate the best evidence with your clinical judgment and the patient's preferences. Present options, discuss risks and benefits, and collaboratively decide on a care plan. Even the strongest evidence may not fit if the patient refuses or if local context makes implementation impossible. STEP 5: ASSESS/EVALUATE—Measure outcomes to determine if the applied evidence improved the situation. Did patient pain decrease? Was infection prevented? Did satisfaction improve? Use the results to refine your approach and contribute to the unit's knowledge base.

The Hierarchy of Evidence ranks research designs and study types by their ability to answer causal questions, from strongest to weakest: LEVEL 1 (STRONGEST)—Systematic reviews and meta-analyses of randomized controlled trials (RCTs). A meta-analysis pools data from multiple RCTs statistically, providing powerful evidence. LEVEL 2—Individual well-designed RCTs. A high-quality RCT with large sample size, low attrition, and rigorous blinding is strong evidence for cause and effect. LEVEL 3—Controlled trials without randomization (quasi-experimental studies). They have control groups but lack random assignment, so causal inference is weaker due to potential bias. LEVEL 4—Cohort studies (prospective, following groups forward) and case-control studies (retrospective, comparing those with and without outcomes). Useful for understanding associations and natural history but weaker for proving causation. LEVEL 5—Systematic reviews of descriptive or qualitative studies. These synthesize qualitative evidence, useful for understanding experience and processes but not for testing causal hypotheses. LEVEL 6—Single descriptive or qualitative studies. A single phenomenology or case study provides rich insight but cannot be generalized. LEVEL 7 (WEAKEST)—Expert opinion, editorials, anecdotal reports. While expert opinion is valuable, it is not grounded in systematic data. When answering clinical questions, always seek evidence from the highest level of the hierarchy possible. If a Level 1 systematic review answers your question, use it. If not available, move down the hierarchy. This hierarchy guides your appraisal step in EBP.

While most staff nurses do not conduct research, you are expected to be an informed, critical consumer of research. Your responsibilities include: (1) Reading current literature relevant to your specialty—stay updated through journals, professional organizations, and unit-based journal clubs. (2) Questioning routines and practices that lack evidence—if 'this is how we've always done it' cannot be justified by research, advocate for change. (3) Identifying and appraising evidence that applies to your patients—when a question arises in practice, search for answers using the EBP process. (4) Participating in data collection when research is conducted in your unit—ensure protocols are followed, protect participant rights, and report problems to the Ethics Review Committee. (5) Implementing research-based guidelines—translate validated findings into safer, more effective care. (6) Contributing to quality improvement initiatives—use data to measure and refine processes. (7) Protecting research participants—understand informed consent, confidentiality, right to withdraw, and report any unethical conduct. Unit-based journal clubs, EBP committees, and adherence to evidence-based clinical guidelines are practical avenues for engagement. In the Philippine context, where resources are constrained and healthcare needs are high, nurses who apply research wisely multiply the impact of limited resources and strengthen the profession's credibility with policymakers.