Nursing Research Methods: Quantitative and Qualitative Sampling, Data Collection, and Statistical Analysis

Comprehensive vocabulary flashcards covering quantitative sampling designs, quantitative data collection and measurement quality, statistical interpretation and hypothesis testing, and qualitative sampling, collection strategies, and analysis traditions.

Sampling

The process of selecting a portion of the population to represent the entire population.

Population

The entire aggregation of cases that meet specific designated criteria.

Sample

A subset of cases selected from a larger population.

Target Population

The entire population of interest that the researcher wishes to study and generalize findings to.

Accessible Population

The cases from the target population that are accessible to the researcher as a pool of subjects.

Eligibility (Inclusion) Criteria

The specific characteristics defined by researchers that delimit and designate who can be in the study population.

Representative Sample

A sample whose key characteristics closely approximate those of the target population.

Sampling Bias

The systematic overrepresentation or underrepresentation of some segment of the population in terms of a characteristic relevant to the research question.

Strata

Mutually exclusive segments of a population established on the basis of a specific characteristic, such as gender or age.

Elements

The individual entities that make up samples and populations (e.g., humans, plants, animals).

Probability Sampling

A sampling design involving random selection where each element has a known, nonzero chance of selection, ensuring that differences between sample and population are purely a function of chance.

Simple Random Sampling

The most basic probability sampling design, wherein elements are selected purely at random from a sampling frame.

Stratified Random Sampling

A probability sampling design in which the population is divided into homogeneous strata from which elements are randomly selected.

Systematic Sampling

A probability sampling design involving the selection of every $k^{\text{th}}$ case from a list or group.

Nonprobability Sampling

A sampling design that selects elements by nonrandom methods, which is rarely representative and cannot guarantee random chance in selection differences.

Convenience Sampling (Quantitative)

Selecting the most conveniently available people as study participants; recognized as the weakest but most commonly used quantitative sampling design.

Example: Nursing student stress

A researcher wants to study stress levels among nursing students.

Instead of randomly selecting students from the entire nursing program, the researcher surveys 30 students who happen to be in the library.

This is convenience sampling because the participants were chosen based on their availability.

Advantages and disadvantages

• Advantages: Quick, easy, inexpensive, and convenient.

• Disadvantages: Higher risk of sampling bias because the sample may not represent the entire population.

Quota Sampling

A nonprobability design where researchers identify population strata and determine the number of participants needed from each stratum, filling quotas using convenience sampling.

Example: Nursing student stress

A researcher wants to study stress levels among 100 nursing students.

Step 1: Divide students into strata.

• Year 1: 400 students

• Year 2: 350 students

• Year 3: 250 students

Step 2: Set a quota for each group.

Strata	Quota
Year 1	40 students
Year 2	35 students
Year 3	25 students
Total	100 students

Step 3: Recruit readily available students from each year until each quota is filled.

Consecutive Sampling

Recruiting all people from an accessible population who meet eligibility criteria over a specific time period or until a specified sample size is reached.

Example: Nursing research

A researcher wants to study pain levels among 100 patients admitted to the emergency department.

Step 1: Establish eligibility criteria (e.g., adult patients experiencing pain).

Step 2: Recruit every patient who meets the criteria and agrees to participate.

Step 3: Continue recruiting until 100 patients are enrolled.

Why is this non-random?

Because only patients who happen to visit that hospital during the study period have an opportunity to participate.

Patients who visit another hospital or arrive after recruitment ends are excluded.

Purposive Sampling (Quantitative)

A nonprobability sampling design where researchers deliberately hand-pick specific cases to be included in the study; considered the strongest nonprobability approach.

Power Analysis

A procedure used to estimate the minimum sample size required to detect an effect of a given size with a specified level of statistical confidence (typically set at $80\%$ power).

Response Rate

The number of people who participate in a study relative to the total number of people sampled.

Nonresponse Bias

Systematic differences between the individuals who agree to participate in a study and those who decline to participate.

Self-Report Data

Data collected via patient-reported outcomes using formal interview schedules or independently completed questionnaires.

Closed-Ended Questions

Questions in which the response alternatives are pre-specified by the researcher, offering ease of administration and analysis.

Dichotomous Question

A closed-ended question format that offers respondents two mutually exclusive answer choices (e.g., 'Yes' or 'No').

Forced-Choice Question

A closed-ended question format requiring respondents to choose between two or more competing statements representing distinct points of view.

Rating Question

A question asking respondents to judge an attribute or experience along an ordered numerical continuum (e.g., from 0 to 10).

Open-Ended Questions

Questions that allow participants to formulate answers in their own words, providing in-depth response data at the expense of ease in analysis.

Interview Schedule

A formal written self-report instrument containing questions asked directly by an interviewer face-to-face or via telephone.

Questionnaire

A self-administered data collection instrument that respondents complete independently on paper or electronically.

Likert Scale

A summated rating scale presenting several declarative statements where respondents indicate their degree of agreement or disagreement along a descriptive continuum.

Visual Analog Scale (VAS)

A straight-line continuum used in self-reports to measure subjective clinical experiences (e.g., pain), where respondents place a mark indicating intensity.

Social Desirability Response Set

A response set bias where participants consistently give answers that reflect socially acceptable values rather than their true opinions or behaviors.

Acquiescence Response Set

A response set bias characterized by a respondent's tendency to agree with statements regardless of content (yea-saying).

Category System

A systematic observational tool with explicit definitions establishing clear criteria for sorting observed phenomena into mutually exclusive categories.

Checklist

An observational instrument used to record the occurrence or frequency of specific pre-listed behaviors.

Time Sampling

An observational sampling technique involving the selection of specific time intervals (systematic or random) during which observations take place.

Event Sampling

An observational sampling strategy that selects specific scheduled or spontaneous events for observation.

Reactivity

A distortion in observational data caused by subjects modifying their behavior when they are aware of being observed.

In Vivo Biophysiologic Measures

Biophysiologic measurements taken directly within or on living organisms (e.g., blood pressure, body temperature).

In Vitro Biophysiologic Measures

Biophysiologic measurements gathered by extracting biological specimens from an organism and assessing them in a laboratory (e.g., hormone levels, bacterial counts).

Reliability

The consistency and absence of variation with which an instrument measures a target attribute upon repeated administrations.

Test-Retest Reliability (Stability)

The degree to which an instrument yields reproducible scores when administered to the same individuals on two separate occasions.

Inter-Rater Reliability

The degree of agreement or consistency between independent observers or scorers using the same measurement instrument.

Internal Consistency

The degree to which all items on an instrument are interrelated and capture the same underlying construct, most widely estimated via Cronbach's $\alpha$.

Reliability Coefficient

A statistical index ranging from $0.00$ to $1.00$ expressing the degree of consistency in measurement, with values below $0.70$ considered unsatisfactory and values $\ge 0.80$ desirable.

Validity

The degree to which an instrument accurately measures what it is intended and supposed to measure.

Face Validity

An appraisal of whether an instrument appears, on the surface and based on subjective judgment, to be measuring the appropriate construct.

Content Validity

The degree to which an instrument adequately samples the full domain of items comprising the construct being measured, frequently assessed by experts using a content validity index.

Criterion Validity

The extent to which scores on a measure correlate with an external gold standard or criterion, requiring a validity coefficient of $r \ge 0.70$.

Construct Validity

The degree to which an instrument provides an accurate operational measure of the theoretical construct of interest rather than an unrelated concept.

Nominal Measurement

The lowest level of measurement, involving the assignment of numbers to label, classify, or categorize attributes without quantitative order (e.g., Married = 1, Single = 2).

Ordinal Measurement

A level of measurement that ranks individuals along an attribute with ordered increments, but does not specify the quantitative distance between ranks (e.g., ADL dependency stages).

Interval Measurement

A continuous level of measurement that ranks subjects on an attribute with known, equal distances between levels, but lacks an absolute, meaningful zero point (e.g., temperature in $^\circ\text{C}$).

Ratio Measurement

The highest level of measurement, possessing equal intervals between units and a true, meaningful zero point that permits statements about absolute magnitude (e.g., height, weight, age).

Parameter

A descriptive characteristic or numerical index summarizing a property of an entire population (e.g., national average).

Statistic

A numerical index or descriptive property calculated from sample data.

Frequency Distribution

A systematic arrangement of numerical values from lowest to highest showing the count ($N$) or percentage ($\%$) of occurrences for each score.

Positive Skew

An asymmetric frequency distribution where the peak is off-center toward the left and the longer tail points toward the right (higher values).

Negative Skew

An asymmetric frequency distribution where the peak is off-center toward the right and the longer tail points toward the left (lower values).

Central Tendency

A statistical index of 'typicalness' that identifies the center or average of a score distribution.

Mean

The arithmetic average of a dataset, calculated by summing all values and dividing by the total number of participants ($\sum x / n$); highly sensitive to extreme outliers.

Median

The exact middle point dividing an ordered set of scores in half ($50^{\text{th}}$ percentile); the preferred measure of central tendency for skewed data.

Mode

The score or numerical value that occurs most frequently in a distribution.

Variability

The extent to which scores in a dataset are dispersed or spread out (heterogeneous indicates high variability; homogeneous indicates low variability).

Range

A measure of dispersion representing the distance between the highest and lowest scores in a distribution ($\text{Range} = \text{highest} - \text{lowest}$).

Variance

A measure of dispersion defined as the average squared deviation of scores from the mean.

Standard Deviation

The square root of the variance ($\sqrt{\text{Variance}}$), indicating score dispersion expressed in original measurement units.

Normal Distribution (Empirical Rule)

A symmetric, bell-shaped distribution in which approximately $68\%$ of values fall within $1\,\text{SD}$ ($\mu \pm 1\sigma$), $95\%$ fall within $2\,\text{SD}$ ($\mu \pm 2\sigma$), and $99.7\%$ fall within $3\,\text{SD}$ ($\mu \pm 3\sigma$).

Cross-Tabulation

A two-dimensional frequency table displaying the joint distribution and relationship between two categorical (nominal or ordinal) variables.

Correlation Coefficient

A statistical index ranging from $-1.0$ to $+1.0$ describing the direction and strength of relationship between two continuous variables (most commonly Pearson's $r$).

Absolute Risk Reduction (ARR)

The absolute numerical difference in the risk of experiencing an adverse outcome between the intervention and control groups.

Odds Ratio (OR)

A risk index representing the ratio of the odds of an outcome occurring in one group compared to the odds of it occurring in another group.

Number Needed to Treat (NNT)

The estimated number of patients who must receive a specific intervention to prevent one additional adverse outcome.

Null Hypothesis ($H_0$)

A statement proposing that there is no true relationship, difference, or effect between independent and dependent variables in the population.

Level of Significance ($\alpha$)

The pre-established probability threshold (typically $p < 0.05$) below which results are judged unlikely to have arisen by chance alone, leading to rejection of the null hypothesis.

Type I Error

The statistical error of rejecting a null hypothesis that is actually true (a false positive; probability equals $\alpha$).

Type II Error

The statistical error of failing to reject a null hypothesis that is actually false (a false negative; probability equals $\beta$).

Confidence Interval (CI)

A computed range of values within which a population parameter is expected to fall at a specified confidence level; if a CI for differences crosses zero, it implies no statistically significant change.

Independent $t$-Test

A bivariate parametric statistical test used to evaluate differences between the means of two independent groups on a continuous dependent variable.

Paired $t$-Test

A bivariate parametric test used to compare the means of a single group under two conditions or time points (e.g., pre-test versus post-test).

One-Way ANOVA

A statistical test ($F$-test) used to compare the means of three or more independent groups on a continuous dependent variable.

Repeated-Measures ANOVA

An analysis of variance test ($F$-test) used to evaluate mean differences across three or more groups or conditions measured over multiple time points.

Chi-Square Test ($\chi^2$)

A nonparametric bivariate test used to assess differences in proportions or relative frequencies across categories between two categorical variables.

Logistic Regression

A multivariate statistical procedure used to model the relationship between two or more independent variables and a nominal (dichotomous) dependent variable.

Multiple Regression

A multivariate statistical procedure used to predict a single continuous dependent variable using two or more independent variables ($R/R^2$).

ANCOVA (Analysis of Covariance)

A statistical procedure used to test differences between group means on a continuous dependent variable while statistically controlling for one or more confounding covariates.

Clinical Significance

The practical importance or meaningfulness of research findings in improving patient care, symptoms, function, or quality of life, independent of statistical significance.

Snowball Sampling (Network Sampling)

A nonprobability qualitative sampling design where early participants refer acquaintances to participate; useful for hidden or vulnerable groups, though restricted to small social networks.

Maximum Variation Sampling

A purposive qualitative sampling strategy that deliberately selects cases exhibiting a wide range of variation along key dimensions of interest.

Typical Case Sampling

A purposive qualitative sampling strategy involving the selection of participants who exemplify what is typical, normal, or average regarding the phenomenon.

Extreme Case Sampling

A purposive qualitative sampling method designed to learn from the most unusual, deviant, or outlier informants (such as outstanding successes or notable failures).

Confirming and Disconfirming Cases

A purposive sampling strategy that selects cases to either support (confirm) or challenge and test (disconfirm) the researcher's emerging theoretical interpretations.

Theoretical Sampling

A grounded theory sampling strategy where decisions about where to find data next are guided by the conceptual needs of the emerging theory.

Data Saturation

The sampling principle in qualitative research dictating that sampling ceases when informational redundancy is achieved and no new themes or properties emerge.

Grand Tour Question

A broad, comprehensive, and open-ended introductory question used in unstructured qualitative interviews to invite participants to narrate their experiences freely.

Topic Guide

A written list of broad questions or topics used by an interviewer to guide discussions in semi-structured qualitative interviews.

Focus Group Interview

A qualitative data collection method involving semi-structured discussions conducted with small groups of 5 to 10 participants and led by a trained moderator.

Photo Elicitation

A qualitative interview technique stimulated, prompted, and guided by photographic images.

Photovoice

A participatory qualitative research method where participants take their own photographs to capture life experiences and interpret them during interviews.