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

Nature and Purpose of Nursing Research

Nursing research is a systematic, rigorous inquiry designed to build, refine, and expand nursing knowledge. Think of it as nursing's way of answering the question: 'Is what we are doing actually working, and how can we do it better?' The purposes of nursing research follow a logical progression: (1) Identify and describe phenomena — simply recognizing and documenting what exists (e.g., describing the prevalence of pressure ulcers in ICU patients in Philippine tertiary hospitals). (2) Explain phenomena — understanding why something happens (e.g., why certain patients develop pressure ulcers faster). (3) Predict outcomes — forecasting what will happen under certain conditions (e.g., predicting which patients are at highest risk). (4) Control phenomena — developing and testing interventions that change outcomes (e.g., testing a repositioning protocol to reduce pressure ulcers). These purposes build on each other progressively. Research also improves care quality, supports clinical decision-making, informs health policy, and strengthens the professional autonomy of nursing as a discipline. In the Philippine context, research competence is embedded in the nursing curriculum by the Commission on Higher Education (CHED) and is reflected in the BON's expected competencies for Registered Nurses.

The Research Process: Steps and Variables

The quantitative research process follows a logical, step-by-step sequence. Think of it as a road map — each step builds on the previous one. Here are the 10 steps: Step 1: Identify and state the problem — Define the gap in knowledge and explain why it matters. The problem must be significant, researchable, feasible, and ethical. Step 2: Review the literature — Systematically search and synthesize existing knowledge. This locates the gap the study will fill and prevents duplication. Step 3: Formulate the theoretical/conceptual framework — The lens or blueprint guiding the study. Theory provides the rationale for how variables are expected to relate (e.g., using Orem's Self-Care Theory to study medication compliance). Step 4: State research questions, objectives, or hypotheses — Specific, measurable statements of what will be studied. Step 5: Select the research design and methodology — The blueprint for HOW the study will be conducted. Step 6: Identify population and sample; select sampling method — WHO will be studied. Step 7: Collect data — Gather information using valid, reliable instruments. Step 8: Analyze data — Apply appropriate statistical tests. Step 9: Interpret findings — Draw conclusions and relate back to the framework and literature. Step 10: Communicate/disseminate and recommend application — Share results through publications, conferences, and policy recommendations. VARIABLES: An Independent Variable (IV) is the presumed cause or the intervention that the researcher manipulates (e.g., a new wound dressing technique). The Dependent Variable (DV) is the outcome or effect that is measured (e.g., wound healing rate). Think: IV causes change in DV. A simple memory aid: IV = Input (what you do); DV = Output (what you measure). HYPOTHESES: The Null Hypothesis (H0) states there is NO relationship or difference between variables (e.g., 'There is no significant difference in wound healing rates between the new dressing technique and standard care'). The Alternative/Research Hypothesis (H1) states an expected relationship or difference. Statistical testing tries to REJECT the null hypothesis. If p < 0.05, you reject H0 and accept H1.

Research Approaches and Designs

Choosing the right research design is like choosing the right tool for a job. The design must match the research question and purpose. There are two major approaches: QUANTITATIVE vs. QUALITATIVE. Quantitative research measures variables numerically, tests hypotheses, and seeks objective, generalizable results using a deductive approach (starting from theory, testing with data). Examples: experiments, surveys with numerical scales, cohort studies. Qualitative research explores meaning, lived experience, and process through narrative, descriptive data using an inductive approach (building theory from data). It answers 'What is it like to experience X?' rather than 'How much of X exists?' There are four major qualitative traditions: (1) Phenomenology — explores the lived experience of a phenomenon (e.g., 'What is it like to be a Filipino nurse working during a typhoon?'). (2) Grounded theory — generates a theory grounded in data from the field (e.g., developing a theory of how nurses make triage decisions). (3) Ethnography — studies the culture and practices of a group (e.g., health beliefs of an indigenous community in Mindanao). (4) Case study — in-depth analysis of a single case or small group (e.g., a detailed study of one hospital's infection control system). QUANTITATIVE DESIGNS: (1) Experimental (True Experiment): Has ALL THREE hallmarks — manipulation of IV, control group, and randomization. The Randomized Controlled Trial (RCT) is the gold standard for establishing cause and effect. It is the strongest evidence for intervention effectiveness. (2) Quasi-experimental: Has manipulation but LACKS randomization and/or a control group. It is weaker for causal inference but more feasible in clinical settings where randomization is unethical or impractical. (3) Non-experimental/Observational Designs: (a) Descriptive — describes characteristics of a phenomenon or population (no manipulation, no hypothesis testing about relationships). (b) Correlational — examines relationships between variables WITHOUT manipulation. Remember: correlation ≠ causation! (c) Cohort (Prospective) — follows a group forward over time to see who develops a condition. Starts with EXPOSURE, ends with OUTCOME. (d) Case-control (Retrospective) — starts with people who have the OUTCOME and looks BACKWARD to find the exposure. (e) Cross-sectional — data collected at ONE point in time; a 'snapshot' study. Good for prevalence studies.

Population and Sampling

You cannot study everyone — that is why sampling exists. The POPULATION is the entire group of interest. It has two components: (1) Target population — the ideal group to which you want to generalize findings (e.g., all Filipino children under 5 years old with acute respiratory infection). (2) Accessible population — the portion of the target population that you can actually reach (e.g., all Filipino children under 5 with ARI admitted to a specific regional hospital in 2024). The SAMPLE is the subset of the accessible population that actually participates in the study. SAMPLING is the process of selecting that subset. A good sample is REPRESENTATIVE — it mirrors the characteristics of the population so that findings can be generalized. SAMPLING ERROR is the difference between the sample results and the true population values. Larger, well-chosen samples reduce sampling error. There are two major categories of sampling: PROBABILITY SAMPLING (supports generalizability): Every member of the population has a known, non-zero chance of being selected. This supports external validity (generalizability). Types: (1) Simple Random Sampling — every member has an EQUAL, INDEPENDENT chance (e.g., drawing names from a hat or using a random number table). Most basic and purest form of probability sampling. (2) Systematic Sampling — every kth member from a list after a random start (e.g., every 5th patient from the hospital census). Simple but may introduce bias if the list has a periodic pattern. (3) Stratified Random Sampling — population is divided into subgroups (strata) based on a key characteristic (e.g., by age group, sex, or province), then random sampling is done within each stratum. Ensures proportional representation of subgroups. (4) Cluster (Multistage) Sampling — groups/clusters are randomly selected first, then individuals are sampled from within those clusters (e.g., randomly select 10 provinces from Luzon, then randomly select hospitals from each province, then randomly select nurses from each hospital). Used for large, geographically dispersed populations. NON-PROBABILITY SAMPLING (weaker generalizability, common in qualitative and preliminary research): Selection is NOT random; not every member has a known chance of being chosen. Types: (1) Convenience Sampling — selecting readily available subjects (e.g., patients in the outpatient clinic on a given day). Most common, most prone to bias. (2) Quota Sampling — convenience sampling with preset quotas for subgroups (e.g., must include 50 males and 50 females). (3) Purposive/Judgmental Sampling — hand-picking participants based on specific characteristics the researcher needs (common in qualitative research). (4) Snowball/Network Sampling — participants refer other eligible participants; used for hard-to-reach groups (e.g., HIV-positive patients, undocumented workers).

Data Collection and Instrument Quality: Validity and Reliability

After selecting the sample, the researcher must COLLECT DATA. Common data collection methods in nursing research include: (1) Questionnaires/Surveys — structured or unstructured; can be self-administered or researcher-administered. (2) Interviews — structured (fixed questions), semi-structured (guided but flexible), or unstructured (open-ended conversation). (3) Observation — direct behavioral observation; can be participant observation (researcher joins the group) or non-participant (researcher observes from outside). (4) Biophysiologic measures — blood pressure, oxygen saturation, blood glucose — objective and precise. (5) Existing records/secondary data — medical charts, hospital databases, vital statistics. The quality of data depends entirely on the quality of the INSTRUMENT (tool) used. Two key properties determine instrument quality: VALIDITY: Does the instrument measure what it INTENDS to measure? Types of validity: (a) Content validity — the instrument covers all relevant aspects of the concept being measured (reviewed by expert panel). (b) Construct validity — the instrument truly measures the theoretical construct it claims to measure. (c) Criterion validity — the instrument correlates with an established gold standard (concurrent = at the same time; predictive = future criterion). RELIABILITY: Does the instrument yield CONSISTENT, REPRODUCIBLE results across time, occasions, or raters? Types of reliability: (a) Test-retest reliability — same instrument, same subjects, different times → should give similar results. (b) Internal consistency — all items in the scale measure the same construct, measured by Cronbach's alpha (acceptable threshold: α ≥ 0.70). (c) Interrater reliability — agreement between two or more observers/raters using the same instrument. CRITICAL RELATIONSHIP: An instrument can be RELIABLE without being VALID (consistently measuring the wrong thing!), BUT an instrument CANNOT be valid without being reliable (if results are inconsistent, it cannot be accurately measuring the right thing either). Think of a bathroom scale that consistently shows you as 5 kg heavier than you are — it is RELIABLE (consistent) but NOT VALID (does not give the true weight). A scale that shows different weights each time you step on it is NEITHER reliable nor valid.

Ethics in Nursing Research

Research ethics protect the rights, dignity, and welfare of human participants. This is non-negotiable in nursing research. The foundational ethical principles come from the Belmont Report (1979) and are expanded in nursing: (1) Respect for Persons (Autonomy) — individuals have the right to decide for themselves whether to participate. (2) Beneficence — do good; maximize benefits to participants and society. (3) Justice — fair distribution of the benefits and burdens of research; do not exploit vulnerable populations. Nursing adds: (4) Nonmaleficence — do no harm; minimize risks. (5) Fidelity — keep promises made to participants. KEY PROTECTIONS: INFORMED CONSENT: The participant's voluntary, informed agreement to participate. Valid informed consent requires: (a) Full disclosure — the participant is told the purpose, procedures, potential risks and benefits, alternatives, and confidentiality protections. (b) Comprehension — the participant understands the information (use language they understand). (c) Voluntariness — participation is COMPLETELY voluntary; no coercion or undue influence. (d) Competence — the participant is mentally capable of giving consent. (e) Right to withdraw — the participant can withdraw at ANY TIME without penalty. CONFIDENTIALITY AND ANONYMITY: Confidentiality means data is kept private and protected from unauthorized access. Anonymity means the participant's identity is NEVER linked to their data (not even by the researcher). In the Philippines, this aligns with the Data Privacy Act of 2012 (RA 10173), which protects personal data. PROTECTION FROM HARM: Risks must be minimized and reasonable in relation to anticipated benefits. The risk-benefit ratio must favor benefits. VULNERABLE GROUPS requiring additional protection: Children, pregnant women, prisoners, cognitively impaired individuals, critically ill patients, and economically disadvantaged persons. For MINORS: TWO levels of consent are required: (1) Parental/guardian consent (because minors cannot legally consent), AND (2) The minor's ASSENT (agreement to participate, appropriate to their developmental level). ETHICS REVIEW: All studies involving human participants must be reviewed and approved by an Institutional Review Board (IRB) or Ethics Review Committee (ERC) BEFORE data collection begins. In the Philippines, research ethics oversight follows the National Ethical Guidelines for Health and Health-Related Research issued by the Philippine Health Research Ethics Board (PHREB) under DOST. The Nuremberg Code and Declaration of Helsinki are also foundational international documents guiding research ethics globally.

Basic Statistics Interpretation

Nurses don't need to be statisticians, but they must interpret statistical results to evaluate research and apply EBP. LEVELS OF MEASUREMENT: Understanding the level of measurement determines which statistical tests are appropriate. (1) Nominal — categories with NO natural order (e.g., sex: male/female; blood type: A, B, AB, O; religion). No mathematical operations meaningful. (2) Ordinal — ordered categories WITHOUT equal intervals between them (e.g., pain scale 0–10; Likert scales: strongly agree to strongly disagree; wound healing stages). Can say one is more than another but not HOW MUCH more. (3) Interval — ordered WITH equal intervals but NO true zero point (e.g., temperature in Celsius/Fahrenheit; IQ scores). A score of 0°C does NOT mean 'no temperature.' (4) Ratio — equal intervals WITH a true zero point (e.g., weight in kg, height in cm, blood pressure, pulse rate, urine output). Zero means 'none.' All mathematical operations are valid. Memory aid: Nominal = Names; Ordinal = Order; Interval = Intervals but no zero; Ratio = Real zero. DESCRIPTIVE STATISTICS — Describe the characteristics of the sample: Measures of Central Tendency: (1) Mean (X̄) — the arithmetic average; sensitive to OUTLIERS (extreme values). Best for normally distributed interval/ratio data. Example: If 5 patients have blood pressures of 120, 130, 125, 128, 127 — Mean = (120+130+125+128+127)/5 = 126 mmHg. (2) Median — the MIDDLE value when data are ordered. BEST for SKEWED data or ORDINAL data (not distorted by outliers). If there is an even number of values, the median is the average of the two middle values. (3) Mode — the MOST FREQUENTLY occurring value. The ONLY appropriate measure of central tendency for NOMINAL data. Measures of Variability: (1) Range — the difference between the highest and lowest values. Simple but unstable. (2) Standard Deviation (SD) — the average distance of each data point from the mean. The most important measure of variability. A higher SD means more spread/variability. THE NORMAL DISTRIBUTION (Bell Curve): In a perfectly normal distribution, Mean = Median = Mode. The data is symmetric. EMPIRICAL RULE (68-95-99.7 Rule): 68% of data falls within ±1 SD of the mean; 95% within ±2 SD; 99.7% within ±3 SD. This is critical for NLE questions! INFERENTIAL STATISTICS — Allow generalization from sample to population and hypothesis testing: p-VALUE: The probability that the observed result occurred by CHANCE (if the null hypothesis were true). Conventional threshold: p < 0.05 = STATISTICALLY SIGNIFICANT → REJECT the null hypothesis. p ≥ 0.05 = NOT statistically significant → FAIL TO REJECT the null hypothesis (do NOT say 'accept' the null). IMPORTANT: Statistical significance ≠ Clinical significance! A result can be statistically significant but too small to matter clinically. TYPE I ERROR (Alpha Error): Rejecting a TRUE null hypothesis = FALSE POSITIVE. Concluding there IS an effect when there ISN'T. The acceptable Type I error rate = alpha level (commonly set at 0.05). TYPE II ERROR (Beta Error): FAILING to reject a FALSE null hypothesis = FALSE NEGATIVE. Concluding there is NO effect when there actually IS. COMMON STATISTICAL TESTS: (1) t-test — compare means of TWO groups (e.g., experimental vs. control). (2) ANOVA (Analysis of Variance) — compare means of THREE or MORE groups. (3) Chi-square test — examine associations between CATEGORICAL (nominal) variables. (4) Correlation coefficient (r) — measure the strength and direction of a LINEAR relationship between two continuous variables. Ranges from -1 to +1. Sign (+/-) indicates DIRECTION; magnitude (how close to 1) indicates STRENGTH. r = 0 means NO linear relationship; r = ±1 means perfect linear relationship. A positive r means as one variable increases, the other also increases. A negative r means as one variable increases, the other decreases.

Evidence-Based Practice (EBP)

Evidence-Based Practice (EBP) is the conscientious, explicit, and judicious use of the BEST AVAILABLE EVIDENCE integrated with CLINICAL EXPERTISE and PATIENT VALUES AND PREFERENCES to make clinical decisions. These three components are the pillars of EBP. You cannot have true EBP if any one of these is missing. RESEARCH vs. EBP vs. QUALITY IMPROVEMENT (QI): Research GENERATES new knowledge through systematic inquiry. EBP APPLIES existing best evidence to clinical practice decisions. Quality Improvement (QI) uses data to improve a SPECIFIC LOCAL PROCESS — not to generate generalizable knowledge. These are distinct activities with different goals. THE 5 A's OF EBP: Step 1 — ASK: Formulate a focused, answerable clinical question using the PICO(T) framework: P = Patient/Population/Problem (who?), I = Intervention (what are you considering doing?), C = Comparison (what is the alternative?), O = Outcome (what do you want to achieve or measure?), (T) = Time frame (optional; over what period?). Example PICO(T): 'In adult post-operative patients (P), does early ambulation within 24 hours (I), compared to ambulation on day 2 or later (C), reduce the incidence of deep vein thrombosis (O) within 30 days post-surgery (T)?' Step 2 — ACQUIRE: Search for the best available evidence using databases (PubMed, CINAHL, Cochrane Library). Apply filters for study design, date, and relevance. Step 3 — APPRAISE: Critically evaluate the evidence for validity (were the methods sound?), importance (is the effect size meaningful?), and applicability (can this be applied to my patients?). Step 4 — APPLY: Integrate the appraised evidence with your clinical expertise and the patient's preferences and values. Implement the evidence-based intervention. Step 5 — ASSESS/EVALUATE: Monitor and evaluate outcomes after applying the evidence. Did it work? Adjust as needed. HIERARCHY OF EVIDENCE (from strongest to weakest): (1) Systematic reviews and meta-analyses of RCTs — STRONGEST: pool data from multiple high-quality RCTs for the most reliable answer. (2) Individual well-designed RCTs — gold standard for single studies. (3) Controlled trials without randomization / quasi-experimental studies. (4) Cohort and case-control studies. (5) Systematic reviews of descriptive/qualitative studies. (6) Single descriptive or qualitative studies. (7) Expert opinion — WEAKEST: subject to individual bias and outdated knowledge. Why does this matter? When choosing evidence to apply, always seek the highest level available for your clinical question.

This question tests the integration of multiple core research concepts. The IV is always what the researcher does or changes. The DV is always what is measured. The design is confirmed by checking all three hallmarks of a true experiment. The null hypothesis always states 'no difference' — it is the default position that statistical testing tries to disprove. Note: The Numeric Pain Rating Scale is an ordinal-level measurement tool.

This problem tests three key statistics concepts: (1) Distribution shape from central tendency relationships — positive skew when mean > median > mode. (2) p-value interpretation — p < 0.05 = significant = reject H0. (3) Types of errors — Type I = false positive (rejecting true H0); Type II = false negative (failing to reject false H0). These are all high-frequency NLE topics.

Systematic sampling is a common NLE topic. The key is knowing: (1) It IS probability sampling. (2) The interval k = population size / desired sample size. (3) Its weakness is periodicity bias. Contrast with simple random sampling (every member has equal, independent chance) and cluster sampling (whole groups are randomly selected first).

The classic NLE trap is temperature in Celsius — students often say Ratio because it has numbers, but 0°C does NOT mean 'no temperature.' The Celsius scale has an arbitrary zero. Ratio requires an ABSOLUTE zero where zero means 'none of the quantity.' Kelvin (K) is ratio; Celsius is interval. Pain scales are ordinal — the difference between a 3 and a 4 is NOT necessarily the same as the difference between a 7 and an 8 in terms of actual pain experience.

This scenario integrates sampling method (snowball), research ethics (confidentiality, voluntariness, vulnerable groups), and Philippine legal context (RA 10173). Snowball sampling is specifically designed for hidden or stigmatized populations. The ethical obligation to protect participants is heightened when the population faces social discrimination — a principle of JUSTICE (fair distribution of research benefits AND burdens, without exploiting vulnerable groups).